DR. P. J. CORKE, APRIL 1995.
This talk will focus on the acute management of patients who have
a spinal cord injury. I will confine the discussion to the initial
management and clinical implications for the anaesthetist involved in
the care of a patient with ASCI in an emergency setting. Spinal
injuries are often a consequence of trauma and the patient may well
have multiple other injuries. I will not be discussing the management
of any associated injury and will assume for the purpose of the
discussion that the injury is confined to the spinal cord. The
intraoperative management of ASCI is a separate topic, as is the
patient with a chronic spinal cord injury, and will not be
discussed.
Acute spinal cord injury (ASCI) occurs at an annual incidence of 20 to 40 per million. In the USA 12,000 new cases per year. Mortality 50%. Young age group affected (15-24 yrs). Incidence of complete ASCI is falling (65% to 45%). This is probable due to improved management. Major causes are MVA's, falls, sports, and recreational injuries. Alcohol and drug abuse are contributing factors. Certain local factors influence the incidence such injuries. Thus, coconuts falling on the heads of labourers are a relatively common cause of ASCI in Malaysia and Singapore. Deliberate damage to the spinal cord, using sharpened bicycle spokes is not an uncommon cause of paraplegia in South Africa. Gunshot injuries are the leading cause of paraplegia in California.
Injuries of the cervical spine are due to:
Hyperflexion.
Usually result from blows to the back of the head or forceful decelerations as might occur in MVA's. They are usually stable and rarely associated with neurological injury.
Hyperflexion-rotation.
Disruption of the posterior ligamentous complex occurs and although cervical nerve root injury is common the spine is stable and not usually associated with spinal cord damage.
Vertical compression or axial loading.
Depending on the magnitude of the compression forces, the resulting injury ranges from loss of vertebral body height with relatively intact margins, to complete disruption of the vertebral body. Posterior displacement of comminuted fragments may result, producing cord injury. Despite cord injury the spine is usually stable.
Hyperextension.
Usually result from a blow to the anterior part of the head or from a whiplash injury. Twice as common as flexion injuries and more often associated with cord damage. Violent hyperextension with fracture of the pedicles of C2 and forward movement of C2 on C3 produces the "Hangman's fracture".
Extension-rotation.
Seen in diving injuries. Because the anterior and posterior columns are disrupted this injury is both unstable and associated with a high incidence of cord dysfunction.
Lateral flexion.
Often associated with extension and flexion injuries.
The mechanism of injury of the thoraco-lumbar spinal cord is similar. The spinal canal is narrower in the thoracic segment relative to the width of the cord, so that when vertebral displacement occurs it is more likely to damage the cord. Until age 10 the spine has increased physiological mobility due to lax ligaments that affords some protection against ASCI. In contrast, elderly patients are at an increased risk due to osteophytes and narrowing of spinal canal.
Anterior 2/3rds of spinal cord supplied by a single anterior spinal artery that arises from the vertebral artery. This circulation includes the corticospinal tracts and motor neurones.
Posterior 1/3rd of the spinal cord supplied by two smaller posterolateral spinal arteries that arise from the vertebral arteries. Occasionally, they arise from the posterior inferior cerebellar arteries. This circulation supplies the dorsal columns.There is essentially no collateral flow between the anterior and posterior circulations.
These spinal arteries are supplemented by segmental (radicular) vessels that arise from intercostal and lumbar arteries (3-5 cervical, 1-2 thoracic, 3-5 lumbar). The large distance between the segmental vessels leaves a watershed area at the upper thoracic and lumbar regions where the cord is particularly at risk of ischaemia.The great radicular artery of Adamkiewicz is of particular importance because it is a source of flow to a large portion of the spinal cord and usually arises from the aorta between T8 and L3. Ischaemia of the cord occurs when the artery of Adamkiewicz is damaged as in aortic dissection. Grey matter blood flow is about 60ml/100g/min and the white matter is about 10ml/100g/min.
A). Spinal cord perfusion pressure (SCPP) = Mean arterial pressure - extrinsic pressure on the spinal cord.
Factors affecting SCBF:
B). Auto-regulation (abolished by trauma). Spinal vessels maintain a constant blood flow within a MAP range of 60-150mmHg by dilating or constricting. If MAP exceeds these limits autoregulation fails and SCBF becomes proportional to MAP.
The primary injury (mechanical injury) results from the original impact and compression against the spinal cord resulting in damage to the small intramedullary vessels causing haemorrhage in the central grey matter and perhaps vasospasm. Primary trauma seldom causes total transection of the cord even though the functional loss may be complete. Primary injury mechanisms include acute compression, impact, missile and distraction forces. The effects of the primary injury depend on the severity and the site. The cervical spine because of its increased mobility is commonly injured after trauma, falls and accidents. The primary injury cannot be treated and can only be prevented with educational programmes aimed at reducing the incidence of ASCI.
The fall in SCBF associated with the primary injury leads to ischaemia of the cord that triggers a biochemical cascade promoting a secondary injury and eventual infarction of the spinal cord and permanent loss of function. Secondary injury occurs within minutes to hours following the primary injury and is thought to involve the release of a myriad of biochemical mediators including:
The long term results of these secondary events are loss of cord conduction and synaptic transmission, membrane damage and cytoskeleton disruption and eventual loss of structural integrity.
Other secondary injury mechanisms include: loss of autoregulation, vasospasm, thrombosis, oedema, hypotension, reduced SCBF and loss of energy metabolism. There is now good evidence that "all is not lost" with the initial primary injury and that the secondary injury is amenable to treatment. I will discuss this later.
A knowledge of the physiological consequences of spinal cord injury will assist the anaesthetist in the management of these patients.
Sudden increase in BP. Bradycardia. Dysrhythmias.
Spinal shock. This results from loss of sympathetic activity below the level of the lesion. The result is physiologically similar to that of spinal anaesthesia and leads to:
Reflex activity may return below the level of the lesion over a period of days to weeks. Somatic or visceral stimulation (eg. bladder or rectal distension) results in massive sympathetic stimulation below the level of the lesion. The intense vasoconstriction increases SVR and BP. There is a compensatory vasodilatation above the level of the lesion associated with flushing, headache, sweating, nasal congestion and pupillary dilation. Ventricular arrhythmias and heart block may be seen. The increase in BP can lead to CVA's, cardiac failure and AMI.
Decreased muscle pump activity, venodilation and pressure on the calves increase the incidence of DVT's and PE's.
The physiological consequences of ASCI on the respiratory system depend on the level of the lesion. The phrenic nerves arise from C3, 4 and 5 nerve roots. Thus if a lesion is below C5 diaphragmatic function is preserved. Injuries above C3 cause instant death unless ventilation is secured immediately. Lesions below C6 cause variable intercostal and abdominal muscle weakness. Intercostal paralysis leads to indrawing of the flaccid intercostal muscles during inspiration. The result is:
The overall effect is severe hypoventilation producing hypercapnia and hypoxaemia. The inability to cough and clear secretions leads to atelectasis and pneumonia. The absence of SNS activity causes reflex bronchoconstriction. The initial surge of sympathetic activity can cause neurogenic pulmonary oedema.
ASCI patients are poikilothermic because of a lack of vasoconstrictors below the level of the lesion. Temperature regulation becomes impaired and they are thus susceptible to hypothermia.
In patients with a partial cord lesion some neurological function is preserved distal to the level of the injury (for example, the sacral segments may be spared). Examination of the whole length of the spine must be performed. An upward movement of the umbilicus on tensing of the abdomen is seen with a T10 lesion (Beavor's sign). Paraplegia describes a state where there is either sensory or motor weakness affecting the two lower limbs. Strictly speaking, anyone with a lesion below T2 will be called a paraplegic. Tetraplegia implies a sensory-motor deficit in all four limbs.
Central cord syndrome: If the zone of injury lies centrally within the cervical cord and encroaches on the long tracts this will produce weakness of the arms. More common in elderly patients (associated cervical spondylosis).
Anterior cord syndrome: Anterior damage affects the corticospinal and spinothalamic tracts. This leads to weakness and impaired pain and temperature sensation. Usually associated with a fracture or dislocation injury.
Posterior cord syndrome: Posterior damage causes loss of vibration and proprioception. More commonly seen after neurosurgical proceedures where a cyst or intraspinal tumour is removed.
Brown-Sequard Syndrome: Trauma may be confined to one side of the cord producing ipsilateral weakness and impaired contralateral pain and temperature sensation ("hemisection of the cord"). More common in penetrating trauma and blunt rotational injury.
Suspect cervical spine injury in any trauma patient and especially if there is any pain or tenderness in the neck. Vertebral injury can occur without cord damage because the spinal canal is widest in the cervical region. Neurological deficit is present in 30-70% of patients with significant spina column injury. Deficits are more common in patients with fracture-dislocations or bone injuries from C5-C7. They range from mild sensory loss to one of several neurological syndromes as previously outlined. Two thirds of all trauma patients have multiple injuries that may interfere with full cervical spine evaluation.
The neck must be aligned in the neutral position and then splinted with a rigid collar. The collar needs to be supplemented by manual stabilisation or lateral support with, for example, sandbags and forehead tape. This decreases movement to about 5% of normal.
Thoracolumbar injury must also be assumed and treated by carefully straightening the trunk and correcting rotation. The patient may be log rolled or lifted as necessary, but it is vital the whole spine is maintained in the neutral position.
Safe transfer of the patient from the stretcher to a trauma trolley can be accomplished by four people, the person in charge grasping the base of the neck and supporting the head with his wrists. In this way there should be no seesaw movements of the neck if the actions of those lifting the patient become incoordinated. Once the patient is lifted the trauma trolley can be wheeled underneath.
The three standard plain views of the cervical spine should be done early ie. cross-table lateral view (CTLV), anterior-posterior and open-mouth views. All seven vertebrae need to be seen if injuries are not to be missed. This can be achieved by applying traction to both arms. C7 fractures account for 20% of cervical spine fractures. As the sensitivity of the CTLV is only 75-80%, a negative X-ray cannot be used for ruling out a C-spine fracture especially if the patient is in a high risk group (head first falls, high speed MVA's, neck pain/tenderness). If the plain films are equivocal then a CT scan is needed. The films should be viewed by a radiologist.
Anteroposterior and lateral radiographs are the standard radiographs of the thoracolumbar spine. Unlike a cervical haematoma a paravertebral haematoma in the thoracolumbar region is best seen on an A-P view in which it may be responsible for mediastinal widening that resembles aortic dissection. The upper thoracic spine is difficult to assess and if signs and symptoms indicate a CT scan should be done. An appreciable force is require to produce an unstable thoracic injury. There will usually be evidence of rib and sternal fractures on the plain CXR.
Obviously the goal is to establish tracheal intubation without causing further injury to the spinal cord. This is a controversial issue and the ultimate method used to secure the airway will depend on the patient's condition, level of cooperation and the skill of the anaesthetist. It should be stated that no method has been shown to be better than the others and the most important factor is to recognise that the spine is unstable and to intubate with great care.
Emergency intubation in an unconscious or uncooperative patient should be achieved under general anaesthesia using a rapid-sequence technique with manual in-line traction (MILT). The patient is immobilised on a long spinal board and one trauma team member minimises spinal movement by controlling the head and limiting neck movement. The front half of the rigid collar should be removed before inducing anaesthesia as it interferes with mouth opening. Thiopentone and suxamethonium are given in appropriate dosage and cricoid pressure is applied in the usual manner. Suxamethonium can be used without danger of hyperkalaemia if the injury occurred within 72 hours. Neck flexion should be avoided and the head extended the minimum amount necessary to allow successful intubation. The overall incidence of neurological complications is about 1-2%. Atlanto-occipital extension occurs during direct laryngoscopy in order to bring the vocal cords into view. Therefore patients with unstable C1 or C2 fractures are most at risk of cord damage during direct larngoscopy. Once intubated skull traction can be applied with various skull callipers. The Gardner-Wells calliper is easily applied and carries a low risk of complications. Halo traction is an alternative. Skull traction helps to correct the alignment of the injured spine, reduce fractures, decompress the cord and nerve roots, and provide stability. Hyperventilation may decrease perfusion and cause ischaemia, therefore normocapnia is recommended.
In a cooperative patient who does not require emergency intubation, awake intubation has been recommended. The fibreoptic technique allows intubation under direct vision with minimal neck movement. The technique requires a sufficiently skilled anaesthetist, a cooperative patient, secretion and blood free airway and adequate topical anaesthesia. Coughing and bucking will result in failure and might threaten an unstable spine. The nasal route should not be used in cases of suspected basal skull fracture. The advantages of this technique are:
Blind nasal intubation is successful in up to 90% of patients but multiple attempts are needed in 67-90% of cases. It can produce nasal haemorrhage and is obviously contraindicated in patients with a coagulopathy.
The airway may be secured via a retrograde catheter technique. Barriot et al used this method in a mobile emergency care unit in 19 patients with cervical spine trauma. All patients were intubated in less than 5 minutes and minor bleeding occurred at the puncture site in 15%.
Cricothyrotomy is usually performed in desperation when all other attempts to secure the airway have failed. There are no studies of neurological outcome after cricothyrotomy in cervical spine injuries.
Physiologically, this is similar to spinal anaesthesia. The resulting vasodilation, hypotension, and if the lesion involves the cardiac accelerator nerves, bradycardia, bradyarrhythmias, and AV block may persist for days to weeks after the initial insult. Bradycardia and hypotension are not classical features of hypovolaemic shock and in a traumatised patient should increase suspicion of spinal cord injury.
Circulatory volume must be restored, but aggressive fluid replacement is detrimental in patients with purely neurogenic hypotension as it precipitates pulmonary oedema (the commonest cause of early death in tetraplegic casualties of the Vietnam war.) A pulmonary artery catheter may be useful especially if the patient has associated injuries or surgery is required. The subclavian route is recommended as access to the internal jugular is difficult without turning the head.The response of the left ventricular filling pressure and cardiac output to incremental fluid challenges can be used to guide fluid therapy. Inability of these patients to increase their heart rate or contractility due to loss of sympathetic output may require vasopressor agents with inotropic properties eg.ephedrine.
Unopposed vagal tone is occasionally severe enough to produce asystole during tracheal suctioning or laryngoscopy and therefore atropine should be given prophylactically.
Intra-arterial pressure monitoring allows beat to beat monitoring of arterial pressure and the rapid assessment of arterial blood gases and haematocrit.
Decreased muscle pump activity, venodilation and pressure on the calves increase the risk of DVT's and PE's. Heparin 5000units subcutaneously twice daily should be commenced and sequential calf compression used intraoperatively.
Paralytic ileus is relatively common after ASCI and may last for 2-3 days. The patient is a risk of gastric aspiration of stomach contents and a distended abdomen can impede respiration. Gastric distension and paralytic ileus require the insertion of a nasogastric tube. An orogastric is used if there is associated basal skull fracture.
Acute retention will develop in paraplegic and tetraplegic patients unless the sacral segments are spared. In the absence of urethral trauma a narrow gauge Silastic catheter is passed under aseptic conditions and taped to the anterior abdominal wall to prevent any unnecessary movement of and injury to the urethra. The urine output should be measured and maintained at greater than 0.5ml/kg/hour.
With loss of the ability to sweat or vasoconstrict within affected dermatomes the patient becomes poikilothermic and needs careful control of his environmental conditions. Core temperature should be monitored and warming devices used to prevent heat loss. Intravenous fluids should be warmed.
Experimental evidence suggests hyperglycaemia aggravates ischaemic injury and dextrose containing infusions should therefore be avoided.
There is now good evidence that "all is not lost" with the initial primary or mechanical injury and that, the spinal cord undergoes additional biochemical and pathological injury, and that the secondary injury is amenable to treatment. Anaesthetists are often involved in the initial resuscitation of patients with ASCI and are therefore in an ideal position to influence the degree of functional recovery that may take place. Traumatic ASCI cause a decrease in local SCBF and loss of autoregulation leading to ischaemia and tissue hypoxia. Superimposed on this background, systemic arterial hypoxaemia and hypotension are not well tolerated by injured neural tissue and may lead to further secondary damage. Hence, the first, and probably most important step in resuscitating the injured spinal cord is to correct hypotension and hypoxaemia.
The exact nature of the secondary injury is unknown but appears to involve post-traumatic ischaemia, continuing compression, oedema, damage by excessive release of free radicals, arachidonate or excitatory amino acids and electrolyte imbalance with intracellular accumulation of calcium and extracellular accumulation of pottasium. The result is infarction of the spinal cord. There are many therapies being investigated both clinically and in the laboratory and I will briefly review these.
Corticosteroids are currently the most widely accepted pharmacological treatment of ASCI. Steroids improve spinal cord blood flow, restore impulse transmission, normalise calcium metabolism and enhance functional neurological recovery following spinal cord trauma in animals. They also inhibit lipid peroxidation, an effect now believed to be its primary therapeutic action in spinal cord injury. The National Acute Spinal Cord Injury Study (NASCIS) in 1990 showed that high dose methylprednisolone (30mg/kg) given within 8 hours of ASCI improved neurological function. However, some patients showed no improvement and no patient recovered neurological function completely. This study has resulted in the widespread clinical use of high-dose steroids in ASCI.
In an attempt to develop more efficacious antioxidants a novel group of compounds were designed, the 21-aminosteroids(lazeroids). They lack glucocorticoid activity and are potent inhibitors of oxygen free radical induced, iron catalysed lipid peroxidation and inhibit the release of free arachidonic acid from injured cells. One compound, tirilazad appears to be particularly promising in cerebral ischaemia and spinal cord injury and is currently in phase III clinical trials for use in head injury.
Based largely on the demonstration that naloxone improves systemic hypotension, SCBF, and neurological recovery in an experimental model of spinal cord injury endogenous opiates were hypothesised to play a role in the evolution of secondary damage following spinal cord trauma. The NASCIS used 5.4mg/kg of naloxone as a bolus followed by 4.0mg/kg/hr and found no better outcome than placebo.
Although the mechanism for the deleterious effect of hyperglycaemia is not entirely clear, one theory is that, in ischaemic areas tissue acidosis from anaerobic metabolism of glucose to lactate is worsened by hyperglycaemia triggering biochemical events such as an increase in intracellular calcium and breakdown of cell membranes that ultimately lead to neuronal death. Thus it seems prudent not to use glucose containing fluids and to maintain normal plasma glucose levels.
Calcium ions play a key role in the pathogenesis of neural injury after trauma and inhibition of calcium ion flux has been a target of treatment. Calcium channel blockers augment cerebral blood flow and neurological recovery after cerebral ischaemia in dogs. Nimodipine improves SCBF immediately after compression injury only if blood pressure is maintained with vasopressors, but fails to improve blood flow and neurological outcome 24 to 96 hours later. It has been proposed that calcium enters the injured neuron by way of channels other than those blocked by nimodipine. The available calcium channel blockers are not currently used in the management of spinal cord injuries.
The free radical scavenger superoxide dimutase reduces the incidence of aortic occlusion induced paraplegia when given intrathecally but is ineffective systemically. Pretreatment with alpha tocopherol and selenium prevent post-traumatic falls in SCBF but their effect on neurological outcome is unknown.
Excitatory amino acids such as glutamate and aspartate appear to be neurotoxic possibly by receptor mediated mechanisms. The postsynaptic receptor known as N-methyl-D-aspartate(NMDA) putatively mediates the neurotoxic effects of glutamate. Hence, MK-801, an antagonist of the NMDA receptor has been studied in models of spinal cord trauma and ischaemia and seems to improve neurological and histological outcome.
This is a complex acidic glycolipid compound present in neuronal membranes. A recent randomised trial of GM1 ganglioside in 34 patients with spinal cord injury demonstrated a modest improvement in muscle strength and overall neurological function.
Adenosine-1 receptors are primarily located in neural tissue, whereas adenosine-2 receptors are found in smooth muscle and endothelial cells. The activation of adenosine-1 receptors decreases neuronal and membrane excitability, thereby limiting the damaging influx of calcium. Aspartate and glutamate release is also inhibited. The infusion of hypothermic adenosine into the aorta of rabbits provides complete protection from spinal cord injury for 40 minutes during aortic cross-clamping. There are no human studies available but animal studies are promising.
Local spinal cord cooling to a temperature of 10°C as therapy was first investigated by Albin et al in 1968. The clinical application, however is impractical since a posterior laminectomy is required to institute the cooling process and it is rarely possible to institute treatment within the 4 to 6 hour post-injury window period during which it may be helpful. On the other hand, reducing rectal temperature to ~34°C is easy to accomplish and has been shown to prolong the duration of ischaemia that the rabbit spinal cord can tolerate without permanent damage.
Since spinal cord ischaemia can be aggravated by systemic hypotension during the spinal shock phase, maintenance or slight elevation of perfusion pressure is beneficial.
1. Grundy D, Swain A, Russel J. ABC of Spinal Cord Injury.
BMJ.1985;292:1558-1560.
2. Tator CH. Acute management of spinal cord injury.
Br.J.Surg.1990;77:485-486.
3. Tator CH, Fehlings MG. Review of the secondary injury theory of
acute spinal cord trauma with emphasis on vascular mechanisms. J
Neurosurg.1991;75:15-26.
4.Young W, Bracken MB. The Second National Acute Spinal Cord Injury
Study. J Neurotrauma1992.Vol 9.Suppl 1.S397-405.
5. Lam AM. Acute spinal cord injury: monitoring and anaesthetic
implications. Can J Anaesth.1991;38:60-67.
6. Bracken MB. A randomised, controlled trial of methylprednisolone
or naloxone in the treatment of acute spinal cord
injury.NEJM1990;322:1405-1411.
7. Swain A, Dove J, Baker H. ABC of Major
Trauma.BMJ.1990;301:34-38.
8. Thiagarajah S. Anaesthetic management of spinal surgery. Anesth
Clinics N Amer.1987;5:587-585.
9. Hastings R, Marks JD. Airway management for trauma patients with
potential cervical spine injuries.Anesth Analg.1991;73:471-82.
10. Crosby ET, Liu A. The adult cervical spine: implications for
airway management.1990;37:77-93.
11. Grande CM, Barton RB, Stene JS. Appropriate techniques for airway
management of emergency patients with suspected spinal cord
injury.Anesth Analg.1988;67:714-15.
12. Suderman VS, Crosby ET, Lui A. Elective oral tracheal intubation
in cervical spine injured adults. Can J Anaesth.1991;38:785-9.
13. Alderson JD, Frost EAM. Spinal cord injuries. Anaesthetic
implications and associated care.1990.Butterworths & Co.
14. Barriot P, Riou B. Retrograde technique for tracheal intubation
in trauma patients. Crit Care Med.1988;16:712-3.
15. Geisler FH, Dorsey FC, Colman WP. Recovery of motor function
after spinal cord injury. A randomised placebo controlled trial with
GM1 ganglioside. NEJM.1991;324:1829-1838.
16. Ford RWJ, Malm DN. Therapeutic trial of hypercarbia and
hypocarbia in acute experimental spinal cord injury. J
Neurosurg.1984;61:925-30.
17. Goto T, Crosby G. Anaesthesia and the spinal cord. Anesth Clin N
Am.1992.10. Vol 3.493-519.
18. Hall ED, Yonkers PA, Andrus PK, Cox JW, Anderson DK. Biochemistry
and pharmacology of lipid antioxidants in acute brain and spinal cord
injury. J Neurotrauma 1992, 9 (suppl 2):425-442.
19. Fraser A, Edmonds-Seal J. Spinal cord injuries.
Anaesthesia.1982;37:1084-1098.
20. Mauney MC, Blackbourne LH, Langenburg SE, Buchanan SA, Kron IL,
Tribble CG. Ann Thorac Surg 1995;59:245-52.
21. Lam A. Spinal cord injury and management. Curr Opinion in
Anaesth. 1992;5:632-639.
22. Meschino A, Devitt JH, Szalai JP, Kich JP, Schartz ML. The saftey
of awake tracheal intubation in cervical spine injury. Can J Anaesth
1992;39:114-117.
23.Albin MS, White RJ, Acosta-Rua G, Yashon D. Study of functional
recovery produced by delayed localized cooling after spinal cord
injury in primates. J Neurosurg.1968;29:113-120.