Chapter 193 Neurologic Complications of Common Spine Operations
Ventral Cervical Surgery
Ventral cervical surgery, whether for discectomy or corpectomy, has many potential neurologic complications, but devastating neurologic injuries are a rarity.1
Specific complication types and rates vary with the side selected for the approach. Injury to the cervical sympathetic chain can occur during dissection of the longus colli muscles from the ventral cervical spine.2 The cervical sympathetic chain ascends on the lateral border of the longus colli muscles and has three ganglionic enlargements. The superior cervical ganglion is at the level of C2-3, the middle ganglion is at the level of C6, and the stellate ganglion is at the level of C7-T1. Injury to the sympathetic chain can be prevented by limiting dissection of the longus colli muscles to their medial aspect only and by careful positioning of the self‑retaining retractor blades on the medial side only of the dissected muscle. The stellate ganglion may occasionally be observed during low dissection in the neck. If it is observed, injury may be avoided by repositioning the retractor to move this structure out of the operating field. Injury to the cervical sympathetic nerves is manifested clinically by Horner syndrome (miosis, anhydrosis, and ptosis). Injury of the cervical sympathetic chain is seldom of any clinical significance, and recovery is usually spontaneous. Injury to the superior laryngeal nerve can occur during ventral cervical discectomy. This usually occurs during dissection in the deep cervical fascia because the superior laryngeal nerve is in close proximity to the superior and inferior thyroidal vessels.3
Injury to the recurrent laryngeal nerve may occur during ventral cervical discectomy, resulting in vocal cord paralysis.2 The nerve is more often injured during a right‑sided operative approach because of the anatomic course of the nerve. In some instances, the right recurrent laryngeal nerve does not loop around the subclavian artery and takes a more direct course, making the nerve more vulnerable to surgical injury. For this reason, some surgeons prefer a left‑sided approach because the left recurrent laryngeal nerve has a more predictable course and has a more medial position between the trachea and the esophagus. However, the left nerve has a higher incidence of idiopathic palsy.3 Vocal cord paralysis from recurrent laryngeal nerve palsy is usually the result of a stretch injury from retraction and usually resolves spontaneously within 6 months after injury. In some patients with vocal cord paralysis, the opposite vocal cord hypertrophies, allowing normal phonation. In any patient who has repeat ventral surgery to the cervical spine, either the approach should be through the previously operated side or the patient should have an otolaryngologic evaluation to confirm normal vocal cord function on both sides before surgery. If vocal cord paralysis is noted, the operative approach should be on the side of the paralyzed vocal cord. Voice hoarseness is often attributable to swelling and does not reflect recurrent laryngeal nerve injury. The incidence of postoperative hoarseness may be lessened by placing a closed suction drain.4 Direct injury to the spinal cord and nerve roots has been reported with an incidence of approximately 2% after ventral cervical operations for myelopathy. These can be avoided only by meticulous operating technique and proper visualization.5
Dural tear is an uncommon intraoperative complication during ventral cervical surgery, more common when the posterior longitudinal ligament is taken down to complete a neural decompression or when corpectomy is carried out. Direct trauma caused by cutting instruments such as the bur or Kerrison rongeurs is the most common cause, but dural defect can be caused by cautery and even by focused application of the bipolar. Dural repair is desirable whenever possible, but access to the site of the leak and consideration for the dangers of retracting and manipulating the cord itself make many repairs unfeasible. In a large series of cervical surgeries that was reviewed for incidence and care of durotomies, laterally located leaks and those occurring behind the vertebral body were found to be unrepairable. Fibrin glue, mobilization precautions, bedrest in a partially seated position, and occasional use of a subarachnoid drain were required to obtain closure in these patients. While successful resolution was inevitably accomplished, additional care and prolonged hospitalization had a significant cost.6
Late postoperative complications of radiculopathy and myelopathy have been reported after ventral cervical discectomy and fusion. Five percent of patients developed myelopathy or radiculopathy an average of 5.5 years after ventral cervical discectomy and fusion, with myelography revealing pathology one level above or below the site of previous fusion. No compression was found at the previously operated site, and only 2.5% of these patients required reoperation at a second level.7 Other researchers have obtained different statistics. Deterioration within the first year after ventral decompression from advancing osteophytic spurring at adjacent levels occurred in 5% of cases, and deterioration in the second and third years after ventral decompressive surgery occurred in another 5% of cases because of osteophytic processes, for a total deterioration rate of 10% within 3 years of surgery.5 This deterioration may be preventable only by postoperative surveillance, reoperating on symptomatic patients, and considering the inclusion of all spondylotic levels in the fusion.5 Complications caused by bone grafts have an incidence of 13%. Nonunion of bone graft was also noted in 7% of patients as an early cause of deterioration and in 4% of patients in the second and third years, for a combined total of 11% for deterioration caused by nonunion.5,8 Nonunion rates may be lowered by paying meticulous attention to bone grafting and by adding ventral cervical instrumentation in multilevel decompressions. Spinal cord compression from malpositioning of the bone graft occurs less often. Bilateral brachial paresis has been reported after ventral cervical surgery for spondylosis. This occurred in a delayed fashion and was associated with angulation at the surgical site, with extrusion of the bone graft.9 Worsening of neurologic function has been reported after ventral cervical fusion from cord compression by the bone graft. Neurologic function improved after removal of the offending graft.5 Proper morticing of the graft bed is key to preventing compression of the spinal cord during graft placement. Precise measurements must be obtained of the depth of the decompression at the superior, inferior, and both lateral walls, as well as the length and width of the graft bed, because these measurements are not uniform. A minimum buffer of 3 mm must be preserved between the decompressed spinal cord and the bone graft, with the mortices cut to provide a dorsal shelf that will prevent the graft from compressing the spinal cord.
Dorsal Cervical Surgery
Central cord syndrome has been reported as a delayed complication that occurs several days after dorsal cervical decompressive laminectomy for cervical stenosis. Central cord syndrome occurred after a period of hypotension and was often associated with abnormal neck position. It is recommended that laminectomy be avoided in patients with abnormal cervical lordosis; that hypotension be avoided, especially when the patient is mobilized for the first time postoperatively; and that a cervical collar be used in the immediate postoperative period.10 Jackson and Simmons11 have reported a case of cervical laminectomy in a markedly kyphotic patient who had local anesthesia, with the patient serving as his own monitor. During the operation, the patient lost neurologic function. This rapidly reversed when the dura was opened. Other reported complications of dorsal cervical surgery include death from air embolism when the sitting position was used, neurologic deterioration from poor positioning of the cervical spine during operation, and tetraplegia from epidural hematoma.12 Neurologic worsening has been noted in myelopathic patients if a laminectomy of inadequate length or width is performed. Laminectomy should extend to the lateral margin of the thecal sac, and if more than 50% of the medial facet is resected, fusion should be considered to lessen the risk of instability. Laminectomy should extend one level higher and one level lower than the highest and lowest compressive lesions.13 Syringomyelia has become symptomatic after cervical decompressive laminectomies for cervical stenosis in patients with an unrecognized syrinx. It is postulated that decompression changes the transmural pressures across the syrinx wall, causing an increase in syrinx size and the appearance of symptoms. This is a rare complication that is lessened by obtaining preoperative MRI to differentiate between syrinx and cervical stenosis.14
Cervical laminectomy may cause instability if more than one half of the medial facet is removed. Other risk factors for postlaminectomy kyphosis include young age and preoperative kyphosis.5 Kyphosis may result in progressive angulation with pain and cervical spondylotic myelopathy. This complication can be lessened by restricting decompression to less than one half of the medial facet or, when more lateral decompression is needed, by performing a fusion.15 During dorsal cervical fusion, spinal cord injury has been reported in a case in which the dorsal bone graft loosened, compressing the spinal cord and causing Brown‑Séquard syndrome.15
Dorsal cervical foraminotomy may worsen radiculopathy.15 Care must be taken not to place any instruments into the narrowed foramen because this will further compress the nerve root, and the nerve root should not be retracted until the foramen is opened, allowing easy retraction of the nerve root.
Thoracic Spine Surgery
Ventral approaches to the thoracic and lumbar spine can cause ischemia to the spinal cord by interrupting the blood supply to the cord. Approaches to the ventral thoracic spine at levels T4-9 are especially dangerous. The risk of causing spinal cord ischemia can be lessened by performing preoperative angiography to localize the artery of Adamkiewicz and then planning surgery to avoid this artery. In addition, temporary segmental artery occlusion with somatosensory-evoked potential (SSEP) monitoring has been advocated to help identify key intersegmental arteries supplying the cord. The segmental arteries are identified and temporarily occluded during a ventral approach to the thoracic spine while SSEP monitoring is carried out. If the SSEP waveform deteriorates, the occlusion is released, and that segmental vessel is spared.16 Thoracic disc herniations may be operated on via a variety of approaches. Neurologic complications are highest with central thoracic disc herniations approached from a dorsal laminectomy. In this setting, the paraplegia rate approaches 18% and is due to excessive spinal cord retraction. Complications may be lessened by a lateral extracavitary approach, costotransversectomy approach, or transthoracic approach.17 Complications may be lessened further by performing preoperative spine angiography to avoid damage to the radiculomedullary artery of the spinal cord.
Lumbar Spine Surgery
Injuries to the dura mater and nerve roots may occur during lumbar disc surgery. Nerve root injury may occur from laceration, thermal injury, and excessive retraction. Lacerations to the nerve root most commonly occur because of lack of identification of the nerve root or because of failure to recognize a flattened root spread over the top of a herniated disc. Adequate illumination and magnification are extremely helpful in locating lumbar nerve roots. The anulus should never be cut until the nerve root has been positively identified. Bone and ligament should be removed until the root can be easily retracted. Further dissection of bone is often needed in the lateral direction to accomplish this goal. Bipolar electrocautery is useful for providing hemostasis, but no cauterization should be attempted until the nerve root is identified to avoid electrical or thermal injury to the nerve root.18 Dural tears may occur during lumbar surgery with an incidence of approximately 4%. Care must be taken to avoid aspirating multiple nerve roots into the suction device and thereby causing neurologic deficit. The dural tear should be covered with a cottonoid, a smaller-diameter suction device should be inserted into the field, and exposure should be improved until the dural tear is fully exposed and then closed, if possible, in a watertight fashion.18
Lumbar discectomy in patients with cauda equina syndrome requires rapid evaluation and treatment. Some authors have found that persistent urinary incontinence is common in patients who are operated on 48 hours or more after presentation of symptoms but is uncommon in those who are operated on within 48 hours.19 Kostuik et al. found two separate modes of presentation in patients with cauda equina syndrome to an acute mode with rapid onset, more severe symptoms, and a poorer prognosis after decompression and a mode with more gradual onset of symptoms. In both groups of patients, those with complete perineal anesthesia tended to have permanent bladder paralysis. Kostuik et al. also found no correlation between timing of surgery and extent of recovery. Despite this lack of correlation, they recommend early surgery for patients with cauda equina syndrome.20
Acute postdiscectomy cauda equina syndrome occurs at a rate of approximately 0.2%.21 The majority of these patients develop symptoms of perineal numbness, urinary retention, motor weakness, and multidermatomal numbness in the recovery room. Stenosis of the lumbar spinal canal at the operative level with anteroposterior dimension of 13 mm or less was the most common factor found in postdiscectomy cauda equina syndrome, with swelling, hematoma, retained disc fragments, and hemostatic gelatin (Gelfoam) contributing to the compression. A large epidural fat graft has also been a reported cause of cauda equina syndrome after lumbar discectomy.22 In this case, a large fat graft herniated into the spinal canal on the first postoperative day, causing cauda equina syndrome. These complications may be avoided by limiting the size of the fat graft to between 5 and 8 mm and by suturing the graft to adjacent paraspinous muscle tissue, by measuring the lumbar canal preoperatively, and, if stenosis is present, by avoiding use of keyhole laminotomy to provide the approach to the disc space. Additionally, hemostasis should be obtained to avoid hematoma formation.23 The operation for lumbar disc herniation causing cauda equina syndrome differs from the usual lumbar discectomy in that a much wider bony exposure is required. Complete hemilaminectomy is recommended to provide space to remove the disc herniation while lessening retraction that could cause permanent deficit. Microdiscectomy should be avoided in these patients because it provides less bony exposure.24
Cauda equina syndrome has also been reported after surgery for lumbar spinal stenosis. Compressive hematoma has been implicated as a cause of cauda equina syndrome after decompressive laminectomy. Cauda equina syndrome has been reported to occur after application of large epidural fat grafts in lumbar decompressive laminectomies.25 This may occur by the fat graft acting to seal a hematoma against the thecal sac or by compression of the large fat graft by the paraspinal muscles. This complication may be prevented by limiting the size of the fat graft to a thickness of 0.5 to 1.0 cm and to a height less than the height of the spinous processes.25 Several cases of cauda equina syndrome have resulted after decompressive laminectomies in which a higher lesion, such as a herniated disc, was later identified. To minimize these complications, the following steps should be taken: Hemostasis should be obtained before closure, or a closed suction drain should be placed if hemostasis cannot be obtained. The entire lumbar spine and thoracolumbar junction should be visualized by MRI or myelography preoperatively so that pathology in the upper portions of the spinal canal is not overlooked. Lumbar decompression should have sufficient length to include all compressed elements. Cauda equina syndrome may also occur on a vascular basis, as the artery of Adamkiewicz may enter at the upper lumbar segments and may be damaged during operation at the thoracolumbar junction. Preoperative spine arteriography should be considered carefully to verify the presence or absence of vascular supply to the spinal cord in the area of the proposed operation if the operation is to be performed in the regions from T4 to L2. In postoperative cauda equina syndrome, mechanical causes must be ruled out and, if present, removed in an urgent fashion. Only after all mechanical causes have been included should a vascular etiology be diagnosed.21
Automated percutaneous discectomy carries risk. The most severe reported complication is one of cauda equina syndrome caused by improper placement of the nucleotome probe in the thecal sac. In animal tests, the device could pierce dura, amputate nerve roots, and make holes in intravascular structures. Complications with the nucleotome probe can be minimized by the following procedures: The nucleotome should never be placed outside the disc space; once properly positioned, the device cannot incise the anulus and therefore cannot exit the disc space. The thecal sac is outlined by the line of the dorsal vertebral bodies ventrally, the line of the junction of the lamina and spinous processes dorsally, and the medial borders of the pedicles laterally; therefore, the probe should be placed under radiographic guidance using these landmarks to avoid the thecal sac. The device should be used only by an operator who has been trained in its proper usage. The procedure should be performed only under local anesthesia, as patient discomfort will alert the operator to any potential injury.26–28
Surgery for lumbosacral spondylolisthesis may also cause cauda equina syndrome even with no attempt at reduction of alignment.29 Paresis of proximal roots after reduction of lumbosacral spondylolisthesis has also been reported and is caused by stretching of proximal nerve roots after reduction of the spine.30 The L5 nerve root is the most often damaged during reduction of lumbosacral spondylolisthesis, followed in frequency by the S1 and S2 nerve roots.30,31 To lessen these complications, SSEP monitoring and prereduction decompression of the nerve roots should be performed. When the patient awakens from surgery, a thorough neurologic examination should be performed on the lumbosacral nerve roots, and if new deficit is found, the patient should be returned to the operating room for decompression of the involved root.29
Surgery at the Cervicomedullary Junction
Operation at the level of the cervicomedullary junction for tumor carries risk. Three approaches are commonly used: the transoral, the dorsolateral, and the lateral suboccipital. The neurologic complications associated with surgery in this area include lower cranial nerve palsy, brainstem infarction from vertebral artery damage, and direct injury to the medulla. These complications may be lessened by the choice of surgical approach. Some surgeons have recommended the lateral suboccipital approach for lesions in this region because it provides direct visualization of the ventral surface of the brainstem and spinal cord and provides a sterile approach with the ability to close the dura in a watertight fashion, and neither mastoidectomy nor transposition of the vertebral artery is required.32
Trauma
Cervical fractures with malalignment may have associated disc herniation. Neurologic deterioration has been reported at a rate of 6 per 68 cases for patients with cervical dislocation undergoing reduction. The preferred management of reduction of cervical dislocation remains controversial. One school of thought advocates closed reduction with emergent ventral cervical discectomy if the patient worsens from disc herniation. Another group recommends MRI to rule out cervical disc herniation before closed reduction to eliminate the risk of neurologic deterioration from herniated disc fragments.33
Gunshot wounds to the spinal cord are the third most common cause of spinal cord injury. Complications of these injuries include osteomyelitis, meningitis, spine instability, cerebrospinal fluid leak, and delayed paraspinal infection. Neurologic recovery is less than 1% in injuries above the T10 level, but as many as 47% of patients with injury to the terminal spinal cord or cauda equina have some degree of recovery with surgical management. Complications may be minimized in patients with transperitoneal gunshot wounds to the spine by vigorous irrigation of the spinal missile track during laparotomy, with rapid closure of any bowel perforations. Debridement and laminectomy are advocated as soon as the patient is stable for gunshot wounds to the spine to provide dural closure, debride bone and missile fragments, and cleanse latent infection.34,35
Neoplastic Disease
Five percent of cancer patients develop neurologic deficits because of spinal metastases.36,37 The majority of metastases occur in the lower thoracic and upper lumbar spine, and 80% of metastases occur ventral to the thecal sac.19 Treatment of these lesions with laminectomy alone often causes instability with resultant paraplegia, especially when there is vertebral body involvement.36 Vertebral body collapse is a particularly ominous finding.38 Treatment in these cases should involve either a ventral approach or a dorsal approach with transpedicular decompression of the ventral thecal sac, reconstruction of the vertebral body with a ventral strut, instrument, or methylmethacrylate and dorsal instrumentation.19,36–39
Spine Instrumentation
Spine instrumentation may cause neurologic injury at the time of implant placement because of neural trauma, ischemia, or hemorrhage or even years after placement from compression by fibrosis or implant loosening. Harrington rods can cause nerve and thecal sac compression in a delayed fashion. Hooks placed on the L5 lamina and attached to Harrington rods have been reported to cut through the lamina over time, causing thecal sac compression. In all these reported cases, the problems resolved after removal of the hardware.40,41
Overdistraction of the spine with spine fixation systems can cause spinal cord injury from ischemia, because distraction stretches the segmental and intrinsic vessels feeding the cord, lessening their cross‑sectional area. Intraoperative SSEP monitoring has been useful but not infallible in detecting overdistraction.42 If SSEP waveforms degrade during distraction, the implant should be removed. If postoperative deficit is noted, even if SSEP waveforms are preserved, consideration should be given to removing the distraction system.42,43 Care must be taken at sites at which the spine is predisposed to injury during distraction. Distraction over a fixed mass, during systolic hypotension, in multicolumn spine injury, and in a kyphotic spine, increases the chance for spinal cord injury.42–44
Malpositioning of pedicle screws can cause radiculopathy, with a reported incidence as high as 7%.45,46 Damage to the nerve root may be caused by the screw itself or by bone fragments from fracture of the pedicle wall. The nerve root travels along the medial border of the pedicle and exits around the inferior border of the pedicle. During placement of pedicle screws, the nerve root is at risk if the medial or inferior border is violated. After the pedicle has been localized and the center of the pedicle has been dissected into the vertebral body, the channel should be probed to detect any breaches into the medial or lateral wall of the pedicle. If no wall defects are found, the channel should be tapped and should be probed after tapping to determine whether any breaches occurred during tapping of the channel. If a defect in the wall is detected, a decision must be made as to whether the channel can be redirected in the same pedicle to avoid the breach or whether an alternative site of fixation must be used. Screw depth should be chosen to penetrate 50% to 75% of the vertebral body, and no attempt should be made to routinely gain bicortical purchase, because neurologic injury to the sympathetic chain in the sacrum and to the presacral plexus at L5 may rarely result. Intraoperative and preoperative imaging with plain films or fluoroscopy is extremely useful for placing pedicle screws.
If a new radiculopathy is noted postoperatively in a patient who has undergone pedicle screw placement at the level of the radiculopathy, radiologic studies should be immediately performed to evaluate screw position. If malpositioning is found, the offending screw should be removed. While placement of the pedicle screw outside of the pedicle wall can cause direct irritation of the nerve root, placement through the neural foramen can result in actual transection or penetration of the nerve or the dorsal root ganglion. The neural function loss and pain experienced in either circumstance can be crippling. Another less common error can occur when the surgeon fails to recognize that the screw is fully inserted and continues to advance the screw down the pedicle. The head of the screw will then advance down the pedicle shaft, trapping and compressing the exiting nerve root under the combination of fractured bone and the base of the screw head as it approaches the back of the vertebral body. Early recognition and revision are the keys to symptom relief if recovery is a possibility.
Ventral cervical fixation devices rarely cause hardware complications that result in neurologic injury. The most common theoretical complication causing neurologic injury is excessive screw length in a ventral cervical plate that compresses or impales the spinal cord. This complication may be avoided by using intraoperative fluoroscopy or plain radiographs while placing these screws. The proper length of screw may be estimated by measuring the adjacent disc space, and the screw should be measured by the surgeon personally before implantation to ensure that the desired screw will be implanted. By using the operative technique described in Chapter 144, this complication may be avoided. Dorsal cervical fixation devices can lead to hardware complications that cause neurologic injury. Dorsal spinous wiring in traumatic injuries to the cervical spine has been associated with worsening of neurologic function by causing retropulsion of herniated disc fragments into the spinal canal.33,47 Nerve root injuries during placement of lateral mass plates are possible and can be avoided by using the technique of placing the screw starting 1 mm medial to the center of the facet and at an angle of 25 to 30 degrees rostrally and 20 degrees laterally to avoid the nerve root and vertebral artery.48
Sublaminar wiring techniques for segmental fixation are associated with a 1% to 17% incidence of neurologic injury.45 Nerve root injuries occur more frequently than do spinal cord injuries with this technique. Acute injuries caused by excessive depth of passage, attempt at passage in an acquired or congenitally narrow canal, and epidural hemorrhage are the most common causes of defects after segmental fixation. Placement of sublaminar wires through the sacral foramina has been noted to cause sacral radiculopathy. Passage of wires medially instead of laterally, thinning of the lamina to lessen length of passage, immediately crimping the wires around the lamina after passage to prevent dislodgement into the canal, and using smaller-gauge double‑twisted wires or cables all decrease the incidence of neurologic injury with sublaminar wiring techniques.9 Removal of sublaminar wires may also cause neurologic injury. The wire should be cut as close to the lamina as possible to minimize the radius of arc that the wire will travel during removal. Single wires and double‑independent wires should be removed individually in a direction parallel to the thecal sac. Double‑twisted wires should be removed with the aid of a wire extractor guide.45
Neurologic complications caused by reducing malalignment of the spine in cases of scoliosis and spondylolisthesis are numerous in the literature. Cases of L5 radiculopathy have occurred during reduction of L5-S1 spondylolisthesis.31 This complication may be avoided by decompression before reduction or by fusion in situ. There are reports of neurologic injury with in situ fusion for spondylolisthesis.29
Surgical Technique
Neurologic complications may be caused by poor surgical technique. The unipolar electrocautery, if used improperly, can cause neurologic damage and should never be used in the vicinity of neural tissue without extreme caution being exercised. Spinal cord damage has been reported from the use of unipolar cautery on the posterior longitudinal ligament.49 Dissection of dorsal structures with electrocautery may result in dissection through the ligamentum flavum and dura with cord or root injury. Care must be used to ensure that dissection is taking place over bone and not over interspace.
High-speed burs and drills can also cause neurologic injury in spine surgery.49 Drills should never be used around unprotected neural structures. Appropriate bit selection is important because a diamond‑tipped drill is less likely to wrap up adjacent tissue than is a fluted bit. The drill should be started near the region of its intended use, and a spinning drill bit should never be moved into or out of the operating field. Direct damage to dura mater or nerve root by the fluted bit occurs, but more commonly, damage occurs when loose soft tissue becomes wrapped up in the spinning bur, either pulling the neural tissue into the bit or pulling the bit into the important tissues. Keeping the working area free of loose fronds of soft tissue is important. Likewise, paddies or other elements that are placed in the canal to protect the dura can become destructive weapons if they become engaged in the bur and are allowed to whip against the neural tissues as the bur spins.
Neurologic damage has also been reported by incorrect identification of anatomic structures. Sectioning of the hypoglossal nerve has been reported in ventral cervical surgery in which the nerve was mistakenly identified as the superior laryngeal artery.49
Allen B.L., Ferguson R.L. Neurologic injuries with the Galveston technique of L‑rod instrumentation for scoliosis. Spine (Phila Pa 1976). 1986;11:14-17.
Boccanera L., Laus M. Cauda equina syndrome following lumbar spinal stenosis surgery. Spine (Phila Pa 1976). 1987;12:712-715.
Cloward R.B. Complications of ventral cervical disc operation and their treatment. Surgery. 1971;69:175-182.
Levy W.J., Dohn D.F., Hardy R.W. Central cord syndrome as a delayed postoperative complication of decompressive laminectomy. Neurosurgery. 1982;11:491-495.
Schoenecker P.L., Cole H.O., Herring J.A., et al. Cauda equina syndrome after in situ arthrodesis for severe spondylolisthesis at the lumbosacral junction. J Bone Joint Surg [Am]. 1990;72:369-377.
Transfeldt E.E., Dendrinos G.K., Bradford D.S. Paresis of proximal lumbar roots after reduction of L5‑S1 spondylolisthesis. Spine (Phila Pa 1976). 1989;14:884-887.
West C.G. Bilateral brachial paresis following ventral decompression for cervical spondylosis. Spine (Phila Pa 1976). 1985;11:176-178.
1. Grisoli F., Grasiani N., Fabrizi A.P., et al. Anterior discectomy without fusion for treatment of cervical lateral soft disc extrusion: a follow‑up of 120 cases. Neurosurgery. 1989;24:853-859.
2. Cloward R.B. Complications of ventral cervical disc operation and their treatment. Surgery. 1971;69:175-182.
3. Carrau R.L., Rivera Cintron F., Astor F. Transcervical approaches to the prevertebral spine. Arch Otolaryngol. 1990;116:1070-1073.
4. Cloward R.B. History of the ventral cervical fusion technique. J Neurosurg. 1985;63:817-818.
5. Yonenobu K., Okada K., Fuji T., et al. Causes of neurologic deterioration following surgical treatment of cervical myelopathy. Spine (Phila Pa 1976). 1986;11:818-823.
6. Hannallah D., Lee J., Khan M., et al. Cerebrospinal fluid leaks following cervical spine surgery. J Bone Joint Surg [Am]. 2008;90:1101-1105.
7. Brandt L., Karlossin M., Holmsted L., et al. Myelography in the late postoperative period in patients subjected to ventral cervical decompression and fusion. Acta Neurochir. 1993;122:97-101.
8. Wiberg J. Effects of surgery on cervical spondylotic myelopathy. Acta Neurochir. 1986;81:113-117.
9. West C.G. Bilateral brachial paresis following ventral decompression for cervical spondylosis. Spine (Phila Pa 1976). 1985;11:176-178.
10. Levy W.J., Dohn D.F., Hardy R.W. Central cord syndrome as a delayed postoperative complication of decompressive laminectomy. Neurosurgery. 1982;11:491-495.
11. Jackson R.P., Simmons E.H. Dural compression as a cause of paraplegia during operative correction of cervical kyphosis in ankylosing spondylitis. Spine (Phila Pa 1976). 1991;16:846-848.
12. Lesion F., Bouasakao N., Clarisse J., et al. Results of surgical treatment of radiculomyelopathy caused by cervical arthrosis based on 1000 operations. Surg Neurol. 1985;23:350-355.
13. Stoops W.I., King R.B. Neural complications of cervical spondylosis and their response to laminectomy and foramenotomy. J Neurosurg. 1962;19:986-999.
14. Middleton T.H., Al‑Mefty O., Harkey L.H., et al. Syringomyelia after decompressive laminectomy for cervical spondylosis. Surg Neurol. 1987;28:458-462.
15. Callahan R.A., Johnson R.M., Margolis J.A., et al. Cervical face fusion for control of instability following laminectomy. J Bone Joint Surg [Am]. 1977;59:991-1002.
16. Apel D.M., Marrero G., King J., et al. Avoiding paraplegia during ventral spinal surgery. Spine (Phila Pa 1976). 1991;16(Suppl):S365-S370.
17. Skubic J.W., Kostuik J.P. Thoracic pain syndromes and thoracic disc herniation. In: Frymoyer J.W., editor. The adult spine: principles and practice. New York: Raven; 1991:1443-1461.
18. Carrol S.E., Wiesel S.W. Neurologic complications and lumbar laminectomy. Clin Orthop Relat Res. 1992;284:14-23.
19. Rompe J.D., Eysel P., Hopf C., et al. Decompression/stabilization of the metastatic spine. Acta Orthop Scand. 1993;64:3-8.
20. Kostuik J.P., Harrington I., Alexander D., et al. Cauda equina syndrome and lumbar disc herniation. J Bone Joint Surg [Br]. 1986;68:386-391.
21. Boccanera L., Laus M. Cauda equina syndrome following lumbar spinal stenosis surgery. Spine (Phila Pa 1976). 1987;12:712-715.
22. Prusick V.R., Lint D.S., Bruder W.J. Cauda equina syndrome as a complication of free epidural fat grafting. J Bone Joint Surg [Am]. 1988;70:1256-1258.
23. Mclaren A.C., Bailey S.I. Cauda equina syndrome: a complication of lumbar discectomy. Clin Orthop Relat Res. 1986;204:143-149.
24. Shapiro S. Cauda equina syndrome secondary to lumbar disc herniation. Neurosurgery. 1993;32:743-747.
25. Mayer P.J., Jacobson F.S. Cauda equina syndrome after surgical treatment of lumbar spinal stenosis with application of free autologous fat graft. J Bone Joint Surg [Am]. 1989;71:1090-1093.
26. Maroon J.C., Onik G., Sternau L. Percutaneous automated discectomy. Clin Orthop Relat Res. 1989;238:64-70.
27. Onik G., Maroon J.C., Jackson R. Cauda equina syndrome secondary to an improperly placed nucleotome probe. Neurosurgery. 1992;30:412-415.
28. Stern M.B. Early experience with percutaneous lateral discectomy. Clin Orthop Relat Res. 1989;238:50-55.
29. Schoenecker P.L., Cole H.O., Herring J.A., et al. Cauda equina syndrome after in situ arthrodesis for severe spondylolisthesis at the lumbosacral junction. J Bone Joint Surg [Am]. 1990;72:369-377.
30. Transfeldt E.E., Dendrinos G.K., Bradford D.S. Paresis of proximal lumbar roots after reduction of L5‑S1 spondylolisthesis. Spine (Phila Pa 1976). 1989;14:884-887.
31. Bradford D.S., Boachie‑Adfei O. Treatment of severe spondylolisthesis by ventral and posterior reduction and stabilization. J Bone Joint Surg [Am]. 1990;72:1060-1066.
32. Pritz M.B. Evaluation and treatment of intradural tumors located ventral to the cervicomedullary junction by a lateral suboccipital approach. Acta Neurochir. 1991;113:74-81.
33. Eisemont F.J., Arena M.J., Green B.A. Extrusion of an intervertebral disc associated with traumatic subluxation or dislocation of cervical facets. J Bone Joint Surg [Am]. 1991;73:1557-1560.
34. Cybulski G.R., Stone J.L., Kant R. Outcome of laminectomy for civilian gunshot injuries of the terminal spinal cord and cauda equina: review of 88 cases. Neurosurgery. 1989;24:392-397.
35. Kihtir T., Ivatury R.R., Simon R., et al. Management of transperitoneal gunshot wounds to the spine. J Trauma. 1991;31:1579-1583.
36. Onimus M., Schraub S., Bertin D., et al. Surgical treatment of vertebral metastasis. Spine (Phila Pa 1976). 1986;11:883-891.
37. Siegal T., Tiqva P., Siegal T. Vertebral body resection for epidural compression by malignant tumors. J Bone Joint Surg [Am]. 1985;67:375-381.
38. Findlay G.F. The role of vertebral body collapse in the management of malignant spinal cord compression. J Neurol Neurosurg Psychiatry. 1987;50:151-154.
39. Galasko C.S. Spinal instability secondary to metastatic cancer. J Bone Joint Surg [Br]. 1991;73:104-108.
40. Hales D.D., Dawson E.G., Delamarter R., et al. Late neurological complications of Harrington‑rod instrumentation. J Bone Joint Surg [Am]. 1989;71:1053-1057.
41. Kornberg M., Herndon W.A., Rechtine G.R. Lumbar nerve root compression at the site of hook insertion. Spine (Phila Pa 1976). 1985;10:853-855.
42. Ginsburg H.H., Shetter A.G., Raudzens P.A. Postoperative paraplegia with preserved intraoperative somatosensory evoked potentials. J Neurosurg. 1985;63:296-300.
43. Johnston C.E., Happel L.T., Norris R., et al. Delayed paraplegia complication sublaminar segmental spinal instrumentation. J Bone Joint Surg [Am]. 1986;68:556-563.
44. Allen B.L., Ferguson R.L. Neurologic injuries with the Galveston technique of L‑rod instrumentation for scoliosis. Spine (Phila Pa 1976). 1986;11:14-17.
45. Chozick B.S., Toselli R. Complications of spinal instrumentation. In: Benzel E.C., editor. Spinal instrumentation. Park Ridge, IL: American Association of Neurological Surgeons; 1994:257-274.
46. McGuire R.A., Amundson G.M. The use of primary internal fixation in spondylolisthesis. Spine (Phila Pa 1976). 1993;18:1662-1672.
47. Abei M., Zuber K., Marchesi D. Treatment of cervical spine injuries with ventral plating. Spine (Phila Pa 1976). 1991;16(Suppl 3):S38-S45.
48. Fehlings M.G., Cooper P.R., Errico T.J. Posterior plates in the management of cervical instability: long term results in 44 patients. J Neurosurg. 1994;81:341-349.
49. Graham J.J. Complications of cervical spine surgery. Spine (Phila Pa 1976). 1989;14:1046-1050.