Chapter 229 Timing of Decompression Surgery for Traumatic Spinal Cord Injury in a Patient with Complete Myelopathy
Presurgical Management
The majority of spinal cord injuries (SCI) each year occur in young patients. Trends are moving toward a greater incidence in older patients, while the over all incidence remains quite stable1,2 (Fig. 229-1). The devastating morbidity and mortality that results from these injuries affects not only the patients, but also their families and society as a whole.3 The early management of SCI is predicated on treating life-threatening injuries as the top priority.2 Advanced Trauma Life Support guidelines support initially treating the greatest physiologic threats, such as respiratory derangements and hypovolemic shock, while providing provisional spine stabilization until definitive treatment can be safely administered. Even in highly unstable injuries that eventually will require complex surgical reconstruction, nonoperative care is the key to initial management.2 The timing of surgical treatment for cervical SCI is still being debated. There are no standards with respect to timing of decompression or definitive stabilization for acute SCI.2 When considering the timing of surgical treatment, one must factor in its potential effects on a host of other important outcome measures, such as neurologic recovery, mortality, hospital length of stay, ICU stay, and infection, to name a few.
Primary and Secondary Spinal Cord Injury
Injury to the spinal cord is the result of a series of mechanisms, including the initial traumatic event and a secondary biologic response to injury.3,4 In the primary mechanism, the energy of the trauma is transferred to the spinal cord, resulting in local deformation, contusion (due to dislocation or encroachment of bony fragments), distraction, or transection/laceration. The spinal cord then suffers secondary injury by a cascade of microvascular, cellular, and molecular events that unfold in the ensuing minutes, hours, and days after the initial SCI.4 Current SCI research that is focused on pharmacologic treatments that might modify this secondary mechanism of injury is beyond the scope of this chapter.
Of the secondary mechanisms of injury that have been identified, persistent compression of the spinal cord represents a principal area of focus that spine surgeons can modify now that might possibly have an effect on patient outcome. Studies have shown that decompression and stabilization of the spine enhance neuronal function and reverse secondary molecular mechanisms of injury.5 While controversy may exist over operative versus nonoperative management and the timing of surgery, there is no debate about the goals of care in the SCI patient. Treatment goals continue to include spine stabilization, realignment, and neurologic decompression. For optimal treatment of the patient, these goals should be addressed at all times throughout the patient’s care from time of injury to definitive management.
Experimental evidence in animals suggests that persistent compression of the spinal cord is a potentially reversible form of secondary injury and that earlier timing of decompression can lead to improved neurologic outcomes.6–8 To date, the best data in support of timing of spinal cord decompression in the setting of SCI have come from studies on animal models of SCI. There is still a lack of consensus on the role of timing of surgical decompression and fixation in SCI patients. Fehlings and Perrin published an extensive review of the literature regarding role and the effect of timing on surgical treatment of SCI. Most studies suggested that surgical decompression and/or stabilization can influence neurologic recovery. Reversal of compression and stabilization of the spine appeared to enhance neuronal function and reverse secondary molecular mechanisms of injury.4 However, a recent systematic review of the literature failed to establish guidelines for timing of surgical decompression and fixation in the setting of SCI due to lack of quality of evidence.3
Despite basic science studies demonstrating that early decompression reverses the secondary injury to the spinal cord and enhances neurologic recovery, historical series suggest that neurologic recovery is possible with nonoperative management as well.9 In these historical studies, the majority of patients were treated nonoperatively. Spontaneous neurologic improvement was reported even in the setting of nonoperative treatment that might not have fully decompressed or realigned the spine.9 Most of these series were retrospective and nonrandomized and suffer from methodologic concerns. However, Frankel et al. reported on a cohort of over 600 patients who were managed nonoperatively. The report showed that almost 30% of the ASIA A injuries improved at least one grade and that neurologic deterioration was rare.9 This landmark study is the standard to which all further studies must be compared. The findings of Frankel et al. were reiterated by many other studies.10–12 In 1981, Kiwerski and Weiss found that 20% of patients with complete SCI showed some neurologic improvement following nonoperative management.10–12 Tator et al. reported on a prospective, nonrandomized case control group of SCI patients to assess outcomes of operative versus nonoperative treatment; they found no difference in neurologic outcome when comparing these two groups but found that operatively treated patients had a lower overall mortality rate.13 Young and Dexter compared two groups of SCI patients and concluded that there is limited hope for recovery from paralysis in those with severe SCI, whether treated operatively or nonoperatively.14
These studies are not meant to support nonoperative treatment over operative treatment in cervical SCI but rather to illustrate the “control” standard to which current interventions must be compared. Current literature suggests that SCIs should be treated operatively.15–18 The basic science studies demonstrating reversal of secondary SCI by early decompression continues to be the foundation that supports many clinicians’ rationale for performing urgent decompression and fixation in patients with SCI.7,8,19 Trends reported over the last decade reveal that the majority of cervical SCIs are being definitively treated surgically.16 Beyond neurology and stability as reasoning for early decompression and fusion, the literature suggests that surgical management of cervical SCIs leads to decreased mortality, ventilation, and time in the ICU.20,21 It is also clear that early surgical stabilization of SCI patients allows for quicker mobilization and transition to rehabilitative efforts, leading to decreased overall hospital length of stay.21
Initial Management (Nonoperative Treatments)
Traction
The application of traction and the timing of its application in the setting of spinal cord compression make up another highly debated area. The role of traction for early indirect reduction, decompression, and stabilization is well founded.22,23 In the Surgical Timing in Acute Spinal Cord Injury (STASCIS) trial, the largest multicenter study on SCI investigating the effect of surgical timing, which is currently ongoing, traction is defined as an acceptable mode of indirect neurologic decompression if meaningful realignment is achieved. Therefore, cervical traction can be thought of as a means of providing rapid spinal cord decompression and temporary stabilization in the early phases of injury.
A number of studies have found no neurologic benefit to early closed reduction of cervical SCI.11,24,25 Fehlings and Perrin concluded in their systematic review that the evidence does not support standards or guidelines for the initial management of SCI with the use of traction and the timing of its application.4 Traction is not without pitfalls.26,27 Tator et al. demonstrated up to an 8% neurologic deterioration rate in patients undergoing traction.26 Others advocate for traction as a necessary means of applying the earliest possible mode of neurologic decompression and advocate immediate application for enhanced chance of neurologic recovery.17,28 Cotler et al. found that patients who underwent cervical spine reduction by traction within the first 8 hours after injury had greater neurologic recovery compared to similar patients who underwent reduction later than 8 hours after the initial injury.28 This timeline was further defined when Aebi et al. reported that patients with cervical dislocations who were reduced less than 6 hours after injury improve to a greater extent than those who were reduced later.17 Brunette and Rockswold and Hadley et al. also provide evidence in studies with small numbers of patients that rapid reduction of traumatic cervical deformity due to fracture can lead to profound neurologic improvement.29,30 These were all retrospective studies, and some provided no comparison groups and are thus limited by their methodology. They still provide useful reference points for assessing treatment guidelines and designing studies.
Kinetic Therapy
In addition to the preservation of life, the main priority of initial SCI treatment is to prevent neurologic deterioration, provide provisional spinal stability, and create the potential for return of neurologic function. Once a patient has been brought under the care of medical personnel, there is still a 6% to 10% risk of neurologic deterioration.27 One of the potential causes of further neurologic deterioration is motion of the spine. The initial form of cervical spine immobilization in the trauma bay or ICU is typically a rigid orthosis or traction and bedrest. However, previous studies have shown that the spine-injured patient who is subjected to bedrest is at risk for pulmonary complications, skin breakdown, deep vein thrombosis, and muscle atrophy.31–35 Frequent patient repositioning has been shown to decrease the morbidity associated with bedrest.32,34,36 However, it also creates the potential for motion of the spine. Kinetic therapy is one way of providing controlled motion to the patient while also maintaining spine immobilization. The various forms of kinetic beds available include Stryker frame (Stryker Corporation, Kalamazoo, MI), RotoRest bed (KCI, San Antonio, TX), CircOlectric Bed (Ferndale Surgical, Inc., Ferndale, MI), and Foster frames.37 Conrad et al. and Rechtine et al. performed biomechanics studies on cadavers with cervical and thoracolumbar instability and compared logroll to kinetic bed treatment. They found that kinetic beds generated significantly less cervical and thoracolumbar spine motion when compared to logroll.38–40 Therefore, kinetic therapy is an excellent mechanism to prevent the complications of immobility while maintaining spine immobility and pursuing potential return of neurologic function.
Patient Positioning in the Operating Room
Prone positioning prior to operative intervention is another potential mechanism of generating further neurologic deterioration due to the generation of motion in the unstable spine.27 The cervical collar does not adequately immobilize the cervical spine when the patient is transitioned from supine to prone positioning, and the manual logroll produces a significant amount of cervical motion that could put neurologic structures at risk. This has been demonstrated in cadavers. In these scenarios, the Jackson table turning technique immobilizes the cervical spine to a significantly greater degree than does manual prone positioning when one focuses on a series of motion parameters for angular displacement (axial rotation, lateral bending and flexion-extension) and linear displacement (anteroposterior, axial, and medial-lateral) whether a rigid collar is used or not.41 The Jackson table prone positioning technique involves positioning a patient supine on the flat Jackson table and then applying the frame, with pads over the ventral surface of the patient. The frame is compressed, and the patient is held in place by the locking pins and compression of the system. Safety straps are applied around the entire setup. The patient is flipped prone by unlocking the manual lock at the head and foot of the bed and then rotating the whole system42 (Fig. 229-2).
Definitive Treatment (Surgery)
Timing of Surgery and Decompression
Although animal studies demonstrate that early cord decompression results in reversal of secondary injury and enhancement of neurologic recovery, the clinical data evaluating the timing of surgical decompression are less clear. Trauma patients often present to SCI centers late and with multiple injuries that confound those findings. According to Tator et al., only 50% of patients with SCI in North America are admitted to SCI centers within 24 hours of their injury.26 This severely limits the patient pool that would even be eligible for what is considered early intervention. In addition, there is not a clear consensus as to the role that timing of spinal cord decompression plays in neurologic recovery. Another significant concern in favor of early operative treatment in patients who present with presumed ASIA A injuries is the difficulty in determining whether the SCI is complete or incomplete in the early phase of injury. It is often challenging to distinguish which patients definitively have complete injuries and which patients are still in spinal shock and therefore might be incomplete (though it is very rare to see a patient who appears to be ASIA A and then, once out of spinal shock, converts to ASIA C or D). Evidence suggests that some patients who were previously identified as having complete SCIs eventually converted to incomplete injuries, occasionally in a delayed fashion.43 Thus, nonoperative management of presumed complete SCIs, as suggested by historical studies, may result in neurologic deterioration if the patient actually had an incomplete SCI. Therefore, some surgeons may choose to treat all cervical SCIs as if they are potentially incomplete (and therefore may decide to operate more quickly) if they weigh the preceding factors more heavily in favor of rapid decompression.
Neurologic Recovery
Duh et al. analyzed data from NASCIS II and reported that patients who underwent surgery less than 25 hours after the initial SCI had neurologic outcomes similar to those of patients who underwent surgery at more than 200 hours after injury.20 Wagner and Chehrazi also showed that timing did not affect neurologic recovery, as they found no difference in neurologic outcome in patients who had surgery within 48 hours of injury compared with those who had surgery after 48 hours from injury.44 Finally, Vaccaro et al. prospectively studied a randomized group of patients with SCI who underwent either early (<72 hours, mean of 1.8 days) or late (>5 days, mean of 16.8 days) surgery.45 There was no significant difference in ASIA motor scores between the two groups at all follow-up time points. Thus, these studies suggest that timing of operative management of complete SCI does not affect neurologic recovery.
However, the study by Mirza et al. provides evidence to the contrary.43 Mirza et al. performed a comparative cohort analysis in which two different centers employed their standard protocols. Members of both institutions performed traction reductions for immediate indirect decompression and stabilization. One of the institutions performed surgery as soon as feasible for the patient, and the other had a protocol of performing delayed surgical fixation 10 to 14 days after injury. Mirza et al. found that the patients in the early-surgery protocol (<72 hours) tended to have higher rates of neurologic recovery during their hospitalization compared to the delayed-surgery group. They hypothesized that surgery provided better stabilization than traction and bedrest. This study was limited by the retrospective nature and nonstandardized treatment algorithms or earlier defined time points for surgery. Also, the study failed to provide data regarding neurologic exams for either patient population after discharge from the acute setting; therefore, it is not known how the long-term neurologic outcomes of these patients compare. Papadopoulos et al. reviewed the effect of treatment timing on 91 nonrandomized patients with acute cervical SCI.15 They found that half of the patients who underwent immediate traction and subsequent surgical decompression had neurologic improvement compared to about one quarter of those who were treated in a nonstandardized manner, typically with delayed surgery. The protocol group had eight patients that recovered from Frankel grade A or B to grade D or E and became ambulatory. These apparent findings of a lower frequency of neurologic improvement in the delayed-treatment group may be confounded by the severity of injury in these patients. Many of the “control” patients had a contraindication to immediate surgery, had other life-threatening injuries that required more immediate care, or were considered futile cases with respect to neurologic recovery.
Non-Neurologic Outcomes
Timing of surgery has been debated not only for its role in neurologic recovery, but also for its role in perioperative and medical complications. Multiple studies have concluded that time of hospitalization and complication rates tend to be related to the severity of injury and multisystem problems.46,47 Many authors have historically argued against surgical treatment (especially early surgery) owing to high risk in critically ill patients.3 Because of improved surgical and anesthetic techniques, more patients are undergoing surgery with little difference in complication rates.2,3
Some studies have found no difference in non-neurologic outcome measures when comparing early and late surgery. Mirza et al. found no difference in major or minor complication rates, ventilator time, and ICU stay between early (mean, 1.8 days) and late (mean, 14 days) surgery groups.43 Vaccaro et al. also found no difference in length of postoperative ICU stay or inpatient rehabilitation time between early and late surgery groups.45 Still other researchers claim that earlier operative treatment helps to improve non-neurologic outcome measures. McKinley et al. reported that early surgery leads to decreased hospital length of stay and reduced pulmonary complications.21 Duh et al. showed that patients who underwent operative treatment less than 24 hours after injury had fewer complications than those who underwent surgery more than 24 hours after injury.20 More recent literature and trends suggest that earlier spine fracture fixation (within 3 days) decreases hospital stay, ventilator days, pneumonia, and hospital costs.16,48–50 Kerwin et al. caution against applying a rigid protocol for early spine stabilization to all patients.48 The timing of surgery should always be individualized to achieve optimization of the patient’s medical and physiologic condition.
Conclusion
There appears to be substantial biologic evidence in the animal literature to support rapid spine stabilization and decompression,3,4 while the relevant timing and patient selection in humans remain unclear.51 It is clear that the goals of treatment for SCI injuries should be to stabilize the spine and decompress the neural elements, but the mechanism and protocols for obtaining these goals are far from rigid. On the basis of current research, there are no clear protocols defining the use of traction and the timing of its application. The same current research appears to indicate that early operative management of spinal cord injuries is safe, but the role for early surgery has yet to be fully defined. It is quite difficult to differentiate a complete and an incomplete SCI in the early phases of injury.43,52 Therefore, some authors recommended treating all SCIs as potentially incomplete SCIs.43,52 Other clinical studies indicate that operative management of SCI leads to improved non-neurologic benefits such as decreased pulmonary complications, time in the ICU, and decreased hospital length of stay.20,21
