Key points

Introduction

Spinal cord injury, affecting approximately 54 per 1 million people annually in the United States, can be a devastating injury for the trauma patient and is associated with a significant morbidity and high mortality rate, particularly among the elderly. Approximately 71.7% of patients with spinal injury suffer polytrauma and the incidence of spinal cord injury among the elderly is rising, making this patient population particularly challenging to manage. Moreover, the socioeconomic burden of spinal cord injury in the United States is substantial. There is a need to optimize treatment paradigms for this patient population. Once patients are resuscitated and stabilized, the cornerstones of management of spinal cord injury include rapid clinical assessment and characterization of injury and, if indicated, definitive surgical decompression and/or stabilization. The surgical approach for decompression and fusion depends on the injury pattern. A critical question faced by the neurosurgeon is the optimal timing for decompression and stabilization. Despite having been studied extensively in the literature, there remains considerable controversy regarding the safety and efficacy of early decompressive surgery. This article examines the evidence for early surgery in patients presenting with spinal cord injury.

Introduction

Spinal cord injury, affecting approximately 54 per 1 million people annually in the United States, can be a devastating injury for the trauma patient and is associated with a significant morbidity and high mortality rate, particularly among the elderly. Approximately 71.7% of patients with spinal injury suffer polytrauma and the incidence of spinal cord injury among the elderly is rising, making this patient population particularly challenging to manage. Moreover, the socioeconomic burden of spinal cord injury in the United States is substantial. There is a need to optimize treatment paradigms for this patient population. Once patients are resuscitated and stabilized, the cornerstones of management of spinal cord injury include rapid clinical assessment and characterization of injury and, if indicated, definitive surgical decompression and/or stabilization. The surgical approach for decompression and fusion depends on the injury pattern. A critical question faced by the neurosurgeon is the optimal timing for decompression and stabilization. Despite having been studied extensively in the literature, there remains considerable controversy regarding the safety and efficacy of early decompressive surgery. This article examines the evidence for early surgery in patients presenting with spinal cord injury.

Pathophysiologic basis for spinal cord injury and experimental evidence for timely decompressive surgery

Blunt traumatic spinal cord injuries, distinct from penetrating injuries, begin with a mechanical insult that results in biomechanical failure of the spinal column leading to bony fractures and/or discoligamentous disruption. Resultant osteoligamentous instability and/or displaced bone fragments can exert compressive, sheer, or distractive forces on the spinal cord itself and can lead to immediate disruption of neural tissue or vasculature. This initial mechanical event constitutes the primary phase of spinal cord injury and the degree of primary injury is correlated with the magnitude of the force of insult. After this inciting event, the secondary phase of injury ensues propagated by vascular ischemia, inflammation, neuronal hyperexcitability, and free radical generation, ultimately leading to further neuronal cell death. Treatment of spinal cord injury focuses on curtailing the extent of secondary injury. Emerging novel therapies target these pathophysiologic processes on a molecular level. Although targeting these secondary processes is appealing and is the focus of intense research, preserving any residual viable neural tissue, even in those patients with complete injury, may optimize the chances of neurologic recovery with these newer treatment modalities.

Persistent compression or forces imparted on the spinal cord through motion from instability are the primary immediate concerns with regard to protecting a patient from further spinal cord injury. Experimental evidence in animal models suggests that the extent of spinal cord injury is correlated both with the degree of cord compression and timing to decompression. Animal studies have demonstrated that a cerebrospinal fluid pressure differential is generated at the level of the compression; this higher pressure cephalad to compression may impair perfusion of viable tissue, leading to ischemia and further secondary injury. Early decompression (<8 hours compared with >72 hours) has been associated with decreased levels of tumor necrosis factor α and fewer apoptotic cells in injured spinal cord tissue, and these factors were associated with improved neurologic recovery. Timely decompression and/or stabilization of spine injuries may preserve remaining viable tissue and reduce the risk of secondary injury.

Initial neurologic assessment after spinal cord injury

Given the neurologic examination may influence the timing of surgical decision making, the initial assessment and management of spinal cord injury are reviewed briefly here. Once a patient with suspected spinal cord injury is identified, a critical first step in management is a thorough neurologic assessment. This allows the neurosurgeon to determine the severity of spinal cord injury and to establish the baseline neurologic status, both of which are important for guiding further management and surgical decision making. Using well-established and validated injury severity grading scales is important because they provide a standardized way to classify spinal cord injury and facilitate communication among practitioners. The American Spinal Injury Association Classifications Standards/International Standards for Neurological Classification of Spinal Cord Injury (ASIA/ISNCSCI) is a widely used and validated clinical grading scale both for the initial evaluation of patients with spinal cord injury and for the assessment of postinjury recovery long term. The ASIA Impairment Scale (AIS) integrates the detailed neurologic assessment captured by the ASIA/ISNCSCI into a simple grading scale of neurologic injury severity ( Table 1 ). The presence of sacral sparing distinguishes a complete injury (ASIA grade A) from incomplete injuries (ASIA grade B–E). A complete spinal cord injury is classified by no evidence of neurologic function, either sensory or motor, at the level of the lower sacral roots (S4-5), indicating the probability of a complete conduction block at the level of injury. On the other hand, the most severe grading of incomplete spinal cord injury, classified as ASIA grade B, requires preservation of some sensory function but no motor function below the level of injury, including the sacral segments. According to the definitions proposed by the International Standards Committee of ASIA, patients with no neurologic function in sacral segments but some retained motor or sensory function elsewhere below the level of injury are still classified as ASIA grade A. Zariffa and colleagues investigated the incidence of this somewhat paradoxic classification in patients with spinal cord injury and found that approximately 3.4% and 34.3% of patients diagnosed as ASIA grade A 1 week from injury would otherwise be designated ASIA grade D and ASIA grade C, respectively, if graded by motor scores alone. Accurate assessment of the lower sacral segments is paramount, because there is clear evidence within the literature of spontaneous recovery of neurologic function in the presence of spared lower sacral nerve root function. In patients in whom lower sacral nerve root function via the anorectal examination is equivocal or unreliable, assessment of sensory function within the high sacral segments is closely correlated and may serve as an adjunct.

Numerous studies have used the ASIA score to categorize injury severity, and there is ample evidence within the literature that more severe spinal cord injury is associated with poor outcome; outcomes are especially poor among those patients with complete injury. There is a clear relationship with AIS grade and functional ambulation, with few patients who are initially ASIA A recovering functional ambulation (see Table 1 ). There is less urgency to definitively treat patients with complete spinal cord injury and well-established surgical grading scales, such as the thoracolumbar injury classification and severity scale and the subaxial cervical spine injury classification system, place less emphasis on operative intervention on patients with complete injury. Hence, distinguishing between an ASIA grade A and ASIA grade B injury may significantly surgical decision making. Although there is ample evidence in the literature for high inter-rater reliability of the ASIA/ISNCSCI among trained professionals, given the reliance on injury severity for operative decision making, it is imperative that an accurate reliable assessment is performed in patients without confounding factors if a neurosurgeon is to use the neurologic examination in the decision making regarding surgery and its timing. Burns and colleagues identified a small but not negligible fraction of patients who were initially diagnosed as ASIA grade A injuries convert to ASIA grade B on follow-up assessment. This is especially important given that 16% to 35% of patients with spinal cord injury have concomitant moderate to severe traumatic brain injury, a confounding factor when attempting a detailed neurologic assessment of spinal cord function. Although the initial neurologic assessment is an important component of surgical decision making, this should not be the sole factor in determining timing of surgery, especially in the setting of confounding factors.

Timing of decompression and neurologic recovery

There is growing evidence within the clinical literature supporting early surgical intervention after traumatic spinal cord injury. The Surgical Timing in Acute Spinal Cord Injury Study (STASCIS) was a multicenter, prospective study of 313 patients with cervical spinal cord injury undergoing surgical decompression and instrumentation. In this study, early surgery was defined as less than 24 hours after the initial injury. The investigators found that even after adjusting for preoperative neurologic status and steroid administration, patients undergoing early surgery were 2.8 times more likely to have at least a 2-grade improvement in AIS score. Patients undergoing early surgery were significantly more likely to be motor complete (ASIA grade A/B; 57.7% vs 38.2%). In another prospective study, Wilson and colleagues examined 55 patients with traumatic spinal cord injury undergoing either early or late decompressive surgery (<24 hours vs >24 hours); after adjusting for presence of complete injury and level of injury, early surgery was predictive of improved neurologic outcomes at discharge from rehabilitation as assessed by the ASIA motor score. Dvorak and colleagues examined changes in ASIA motor score in patients with incomplete spinal cord injury and found significantly greater degree of neurologic recovery in patients undergoing early surgery (a 6-point motor difference). Although a 6-point motor score difference may not seem substantial, such an improvement may lead to functional recovery of 1 or 2 neurologic levels that may translate into greater independence and quality of life. For example, patients who convert to a C7 from a C6 neurologic level regain the arm extension and the ability to perform independent transfers. Dvorak and colleagues did not observe improvement with early surgery in patients with complete spinal cord injury.

As suggested previously, in the subgroup of patients with complete spinal cord injury, the timing of decompressive surgery is even more controversial. In general, outcomes within this subgroup of patients are poor, although few studies have closely examined patients with complete spinal cord injury. Recently, Bourassa-Moreau and colleagues specifically studied patients with complete spinal cord injury, defined as AIS grade A on initial assessment who underwent surgery either in an early or delayed fashion (less than or greater than 24 hours after injury, respectively). Although the observed frequency of neurologic improvement with early surgery did not reach statistical significance overall, the investigators noted that among patients with cervical injuries, AIS grade A patients had significantly greater rate of neurologic improvement in the early surgery group when compared with the delayed surgery group (64% vs 0%, respectively). There are several studies that contradict this finding, showing no benefit in neurologic recovery in patients with early versus later surgery.

Not all studies support early surgical intervention. Vaccaro and colleagues conducted a prospective study of 62 patients examining surgery performed within 72 hours or greater than 5 days after cervical spinal cord injury. In this early clinical study, the investigators did not find any difference in neurologic outcomes at follow-up. In a retrospective study of 595 patients with cervical spinal cord injury, patients who underwent surgery either 72 hours before or after the initial injury were assessed for neurologic improvement. In this study, neurologic improvement was not significantly different between the surgical groups; patients undergoing early surgery experienced greater degrees of postoperative neurologic decline and perioperative mortality. In another retrospective review, Pollard and colleagues studied 412 patients with incomplete cervical spinal cord injury and found no relationship between early surgery (<24 hours postinjury) and late surgery.

Given the inconsistent findings in the literature, several meta-analyses have been published to address the issue of timing of surgical decompression. A recent meta-analysis of published studies by Liu and colleagues found that patients undergoing surgery within 24 hours of injury had greater degrees of neurologic improvement and overall motor function. In another meta-analysis, van Middendorp and colleagues found a similar benefit in the early surgery group; the investigators, however, found significant evidence of publication bias secondary to study heterogeneity and advised readers to interpret their results with caution.

More recently, the impact of even earlier decompressive surgery within 8 hours of injury has been investigated. Grassner and colleagues performed a retrospective analysis of patients undergoing decompressive surgery within 8 hours or after 8 hours of injury; in addition to baseline neurologic status and age, timing of surgery independently predicted neurologic improvement at follow-up as assed by change in Spinal Cord Independence Measure. In another study, Jug and colleagues examined patients who underwent early decompressive surgery dichotomized by intervention less than 8 hours from injury and between 8 hours and 24 hours from injury; even within this group of patients undergoing early surgery, decompression less than 8 hours from injury was associated with a significantly increased odds of a 2-grade improvement in AIS score even after controlling for baseline neurologic function and amount of canal compromise. There was no difference in perioperative complications or mortality in these 2 groups, suggesting that surgical intervention can be performed very early after injury without added perioperative risk.

In summary, there is accumulating evidence within the literature regarding the neurologic benefit of early surgery after spinal cord injury. In particular, there seems to be a more robust impact on neurologic recovery with early surgery in patients with incomplete injury, and there are data that the earlier the surgery, the greater the neurologic benefit. Moreover, there is emerging evidence that even in patients with complete spinal cord injury, early intervention may yield neurologic improvement. This topic remains controversial given the overall small sample, the heterogeneity of published studies, and lack of level I data. Randomized clinical trials examining timing of surgery are unfeasible due to obvious ethical issues. Moreover, definitions of early surgery within published studies are varied. Larger, multicenter studies that are adequately powered are needed to (1) identify the optimal definition of early surgery, (2) elucidate the true role of decompression after neurologically complete injuries, and (3) delineate patient factors (ie, specific level of injury) that are predictive of improved neurologic recovery and functional outcome with early surgery.