Spinal Cord Ischemia
Also known as: spinal cord injury, postoperative paraplegia, postoperative paraparesis, thoracoabdominal aortic aneurysm repair complication
1. Description of the problem
What every clinician needs to know
Spinal cord ischemia is a relatively uncommon form of spinal cord injury. The true prevalence of cord ischemia is not known, but it is suggested that fewer than 2% of central neurovascular events affect the cord and fewer than 8% of all acute myelopathies have an ischemic component. The infrequent nature of spinal cord ischemia is due to the multiple collateral sources of blood supply. Rostrally, branches from the vertebral arteries come together in the midline to form the anterior spinal artery. Caudally, blood to the anterior cord is provided by radicular arteries which arise from the aorta at multiple levels and anastomose with dural branches posteriorly.
This major source of blood flow is well protected and makes cervical ischemia a rare event. The lumbosacral region also has large perforating branches; the best known is the artery of Adamkiewicz, which is usually found between T8 and L3, making the lumbar region fairly resistant to ischemia as well.
Despite the highly vascular nature of the cord, there are several conditions that can result is reduced blood flow to the cord. The most common causes for cord ischemia include global hypotension, thoracoabdominal surgery and mechanical cord compression. Global hypotension is likely underappreciated due to the high degree of other end-organ injury and dysfunction that can mask the cord injury. It is estimated that as many as 45% of hypoxic ischemic encephalopathy cases also have cord ischemia, though this is often masked by the brain injury.
In operative cases, open and endovascular surgical procedures (e.g. descending thoracic and thoracoabdominal aortic aneurysm repair) may result in permanent parapegia in 1-5% of cases but may cause some degree of permanent neurologic deficit in up to 14% in one series. In some patients, cervical stenosis can predispose to cord ischemia under conditions of neck extension. Similarly, hemorrhage into the cord or around the cord may cause compression of the arterial supply and produce cord ischemia. Less common causes are prior radiation therapy, which can lead to myelopathy and progressive ischemia and embolic phenomena. These causes are summarized in
There is no classic picture of cord ischemia; however, there are a few aspects of the presentation that can suggest that ischemia has a major part in the presentation of spinal cord dysfunction. One of these aspects is the prominent feature pain has in the presentation; with weakness typically following by minutes to hours later. The pain is often in a radicular pattern. Urinary retention is also common, as is seen with other acute cord injuries. Ultimately the weakness is the most consistent feature, often greater than the sensory deficits.
Recovery is quite variable, though the more severe the presentation the worse the recovery. Further, the less motor recovery seen in the first 24 hours, the less likely significant recovery will be seen. Management is limited to lumbar drain placement to reduce pressure on venous outflow and to promote perfusion, and to the use of pressers to elevate perfusion pressures to the cord. Ischemia of the cord can be confirmed by MRI with diffusion weighted sequences.
2. Emergency Management
1. Avoid or correct any hypotension.
a. Maintain MAP above 65mm Hg.
b. Unless specifically contraindicated, treat with IV fluids and vasopressor agents for a MAP of greater than 85mm Hg for 24 hours or more.
2. If spinal ischemia has occurred after thoracoabdominal surgery, consider placement of a lumbar drain for therapeutic drainage of cerebrospinal fluid. There are no standard drainage guidelines, but consider 10-15 ml/hour or titrate to neurologic exam if possible.
3. If spinal cord ischemia has resulted from a known or suspected embolic origin, consider IV tissue plasminogen activator (tPA).
a. Administer according to standard exclusion and inclusion criteria.
b. If embolic source is known, consider interventional neuroradiology-guided embolectomy or intra-arterial thrombolytic administration.
4. During the first 24-48 hours of treatment, maintain normothermia with standard hyperthermia treatment maneuvers.
Diagnosis of spinal cord ischemia will be based primarily on history and physical examination rather than specific imaging techniques. However, with certain software, MRI with diffusion weighted sequences can be helpful (Figure 1).
Similarly, in cases of spinal cord ischemia as a complication of endovascular aortic repair, angiography can diagnose the arterial occlusion. In iatrogenic cases, the proximate cause of the injury will often be obvious, particularly after a prolonged thoracoabdominal aorta repair case.
4. Specific Treatment
Once the diagnosis has been made, the first step in management is to maintain an adequate and often elevated perfusion pressure to the spinal cord. In most cases, the targeted pressure is a mean arterial pressure (MAP) of 85mm Hg. However, the most important factor is ensuring that no hypotension occurs. as this will invariably worsen the patient’s neurologic exam.
In some cases, particularly those with spinal cord ischemia after thoracoabdominal aorta surgery, practitioners will titrate the MAP to the patient’s best physical exam. Both fluids and vasopressors such as phenylephrine or norepinephrine are often necessary to maintain an elevated MAP. Most practitioners will continue these goals for at least 24 hours. However, often the support is weaned and if the exam worsens, the interventions are continued for another 24 hours.
In cases of suspected acute spinal cord stroke, tPA can be considered. There are a few case series of patients treated with IV thrombolytics after known or suspected stroke, with variable results. If such therapy were to be initiated, the authors would suggest adhering to the standard inclusion and exclusion criteria for IV tPA in acute cerebral stroke.
Additional therapies that can be considered include intravenous steroids. As with other treatments for central nervous system ischemia or injury, this therapy should be approached with caution. In traumatic spinal cord injury, there may be a mild benefit with corticosteroids, though infection risk is elevated. However, in ischemic and hemorrhagic intracranial stroke or trauma-related edema, corticosteroid therapy is contraindicated, as there is a worsened long-term outcome.
Physical therapy should begin as soon as possible. If surgical stabilization is required or if drains and pressers are being used, rehabilitation therapy should be initiated as soon as these are weaned. Continued range of motion can be done while in bed.
The use of venous thrombosis (VT) prophylaxis also is critical. Mechanical intervention with sequential compression devices should be started immediately, with chemical prophylaxis added as soon as medically safe. If hemorrhage is part of the clinical presentation then prophylaxis may be held for 5-7 days.
When drains are used, heparin can be used 24 hours after the drain is placed and until 12 hours prior to the drain being pulled. In all other cases, and following drain removal, Lovenox is the preferred agent for VT prevention. Typically doses of 40mg subcutaneously once daily or 30mg subcutaneously twice daily are used. Prophylaxis with Lovenox or anticoagulation should continue for 6 months and then can be discontinued.
5. Disease monitoring, follow-up and disposition
Expected response to treatment
Typically, the suggested interventions should result in an improvement in the exam. There are no strict durations for the elevated MAP goals or CSF drainage, though generally 3-5 days is done. When exam improves, pressors and drainage are weaned and the exam is followed closely. If the exam worsens, the interventions are continued. In one series, absence of motor improvement in the first 24 hours was predictive of poor outcome long-term. If, at its peak, the motor deficits are not severe, nearly 50% of patients regain the ability to walk, according one prospective study.
In all cases of acute spinal cord dysfunction, a neurosurgery or orthopedics consult should be obtained to allow for rapid surgical intervention if symptoms progress or new problems arise during management. Repeat imaging is always indicated for evaluation of patients with worsening exams. MRI remains the imaging study of choice for the first and any subsequent imaging.
Physical therapy should begin as soon as possible. If surgical stabilization is required or if drains and pressors are being used, then as soon as these are weaned rehabilitation therapy should be initiated. Continued range of motion can be done while in bed. Critical too is the use of venous thrombosis prophylaxis. Mechanical intervention with sequential compression devices should be done immediately, with chemical prophylaxis added as soon as medically safe. If hemorrhage is part of the clinical presentation then prophylaxis may be held for 5-7 days.
In a general sense, the underlying problem is blood flow to the spinal cord. There are several mechanisms that can cause a reduction in flow and these were reviewed elsewhere. Here we provide a more detailed table of symptoms related to the various artery levels that are affected. Regardless of the cause, as summarized in Table 2, the ultimate result is a reduction in blood flow to the cord that results in ischemic injury.
As mentioned elsewhere, the true incidence of spinal cord ischemia is unknown, though autopsy studies indicate that as many as 45% of hypoxic ischemic cases, up to 8% of acute myelopathies and about 1-2% of neurovascular events have associated cord ischemia.
In one series, absence of motor improvement in the first 24 hours was predictive of poor outcome long-term. If, at its peak, the motor deficits are not severe, nearly 50% of patients regain the ability to walk according one prospective study. In cases that are moderate severity, about 30% will still require a wheelchair at 4 years.
What's the evidence?
Geldmacher, DS, Shah, L. “Vascular Myelopathies”. Continuum. vol. 14. June 2008. pp. 71-90. (This entire Continuum issue is dedicated to reviewing spinal cord, root and plexus disorders. Great, concise comprehensive resource for this material. The section cited above specifically addresses spinal cord ischemia. Note that both tables in this chapter were derived from the tables provided in this reference.)
Sandson, TA, Friedman, JH. “Report of 8 cases and review of the literature”. Medicine (Baltimore). vol. 68. 1989. pp. 282-292. (A small case series of spinal cord infarction collected over 4 years in a community setting accounting for 1% of stroke cases.)
Causes of Cord Ischemia
Novy, J, Carruzzo, A, Maeder, P, Bogousslavsky, J. “Spinal cord ischemia: clinical and imaging patterns, pathogenesis and outcomes in 27 patients”. Arch Neurol. vol. 63. 2006. pp. 1113-20. (A well done retrospective case series and detailed review of spinal cord ischemia describing the two main subtypes of spinal cord ischemia distribution.)
Cheshire, WP, Santos, CC, Massey, EW, Howard, JF. “Spinal Cord Infarction: etiology and outcome”. Neurology. vol. 47. 1996. pp. 321-30. (A large retrospective review of spinal cord ischemia and infarction from two university hospitals, including analysis of injury distribution and outcome of patients.
Duggal, N, Lach, B. “"Selective vulnerability of the lumbosacral spinal cord after cardiac arrest and hypotension,"”. Stroke. vol. 33. 2002. pp. 116-21. (A large retrospective review of spinal cord ischemia after global ischemic events indicating selective vulnerability of the lumbosacral spinal cord.)
Gass, A. “MRI of Spinal Cord Infarction”. Neurology. vol. 54. 2000. pp. 2195(A brief case description of spinal cord infarction with image. Reference for image uploaded.)
Cheung, AT, Pochettino, A, McGarvey, ML. “Strategies to manage paraplegia risk after endovascular stent repair of descending thoracic aortic aneurysm”. Surg. vol. 80. 2005. pp. 1280-88. (This paper describes a plan for surveillance and treatment of spinal cord ischemia after placement of thoracic aorta stents including induced elevation of mean arterial pressure and placement of lumbar drains for CSF diversion.)
Venous thrombosis prevention
Gould, MK, Garcia, DA, Wren, SM, Karanicolas, PJ, Arcelus, JI. “Prevention of VTE in nonorthopedic surgical patients: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines”. Chest. vol. 141. 2012 Feb. pp. e227S-77S. (ACCP guidelines on venous thromboembolism prophylaxis in patients with stroke and other non-orthopedic patients.)
Masson, C, Pruvo, JP, Meder, JF. “Spinal cord infarction: clinical and magnetic resonance imaging findings and short-term outcome”. J Neurol Neurosur Psychiatry. vol. 75. 2004. pp. 1431-5. (This paper describes a prospective cohort of patients with spinal cord infarction, including imaging characteristics and prognostic factors.)
Nedeltchev, K, Loher, TJ, Stepper, F. “Long-term outcome of acute spinal cord ischemia syndrome”. Stroke. vol. 35. 2004. pp. 560-5. (A well done retrospective analysis of 57 patients with spinal cord ischemia identifying important prognostic indicators.)
Pelser, H, van Gijn, J. “Spinal infarction. A follow-up study”. Stroke. vol. 24. 1993. pp. 896-8. (A small case series following the long-term outcome of patients with spinal cord infarction that shows improvement in motor function but with significant disabling pain.)
Copyright © 2017, 2013 Decision Support in Medicine, LLC. All rights reserved.
No sponsor or advertiser has participated in, approved or paid for the content provided by Decision Support in Medicine LLC. The Licensed Content is the property of and copyrighted by DSM.
- Spinal Cord Ischemia
- 1. Description of the problem
- 2. Emergency Management
- 3. Diagnosis
- 4. Specific Treatment
- 5. Disease monitoring, follow-up and disposition
- What's the evidence?