Early time-dependent decompression for spinal cord injury: vascular mechanisms of recovery.

Although surgical decompression is often advocated for acute spinal cord injury, the timing and efficacy of early treatment have not been clinically proven. Our objectives were to determine the importance of early spinal cord decompression on recovery of evoked potential conduction under precision loading conditions and to determine if regional vascular mechanisms could be linked to electrophysiologic recovery. Twenty-one mature beagles were anesthetized and mechanically ventilated to maintain normal respiratory and acid-base balance. Somatosensory-evoked potentials from the upper and lower extremities were measured at regular intervals. The spinal cord at T-13 was loaded dorsally under precision loading conditions until evoked potential amplitudes had been reduced by 50%. At this functional endpoint, spinal cord displacement was maintained for either 30 (n = 7), 60 (n = 8), or 180 min (n = 6). Spinal cord decompression was followed by a 3-h monitoring period. Regional spinal cord blood flow was measured with fluorescent microspheres at baseline (following laminectomy) immediately after stopping dynamic cord compression, 5, 15, and 180 min after decompression. Within 5 min after stopping dynamic compression, evoked potential signals were absent in all dogs. We observed somatosensory-evoked potential recovery in 6 of 7 dogs in the 30-min compression group, 5 of 8 dogs in the 60-min compression group, and 0 of 6 dogs in the 180-min compression group. Recovery in the 30- and 60-min groups varied significantly from the 180-min group (p < 0.05). Regional spinal cord blood flow at baseline, 21.4+/-2.2 ml/100/g/min (combined group mean +/- SE) decreased to 4.1+/-0.7 ml/100 g/min after stopping dynamic compression. Reperfusion flows after decompression were inversely related to duration of compression. Of the 7 dogs in the 30 min compression group, 5 min after decompression the blood flow was 49.1+/-3.1 ml/100 g/min, which was greater than two times baseline. In the 180-min compression group early post-decompression blood flow, 19.8+/-6.2 ml/100 g/min, was not significantly different than baseline. Of the 8 dogs in the 60-min compression group, 5 who recovered evoked potential conduction revealed a lower spinal cord blood flow sampled immediately after stopping dynamic compression, 2.1+/-0.4 ml/100 g/min, compared to the 3 who did not recover where blood flow was 8.4+/-2.1 ml/100 g/min (p < 0.05). Reperfusion flows measured as the interval change in blood flow between the time dynamic compression was stopped to 5, 15, or 180 min after decompression, were significantly greater in those dogs that recovered evoked potential function (p < 0.05). Three hours after decompression, spinal cord blood flow in the 3 dogs in the 60-min compression group with no recovery, 11.1+/-2.1 ml/100 g/min, was significantly less than the spinal cord blood flow of the recovered group (n = 5), 20.5+/-2.2 ml/100 g/min. These data illustrate the importance of early time-dependent events following precision dynamic spinal cord loading and sustained compression conditions. Spinal cord decompression performed within 1 h of evoked potential loss resulted in significant electrophysiologic recovery after 3 h of monitoring. This study showed that the degree of early reperfusion hyperemia after decompression was inversely proportional to the duration of spinal cord compression and proportional to electrophysiologic recovery. Residual blood flow during the sustained compression period was significantly higher in those dogs that did not recover evoked potential function after decompression suggesting a reperfusion injury. These results indicate that, after precise dynamic spinal cord loading to a point of functional conduction deficit (50% decline in evoked potential amplitude), a critical time period exists where intervention in the form of early spinal cord decompression can lead to effective recovery of electrophysiologic function in the 1- to 3-h post-decompression p

Demonstration of diastematomyelia and associated abnormalities with MR imaging.

Three patients were studied with a 0.3 T superconducting magnet to assess the role of magnetic resonance (MR) imaging in the recognition and evaluation of diastematomyelia and associated abnormalities. Comparison was made with other imaging techniques, including metrizamide computed tomographic (CT) myelography. With MR imaging, the divided spinal cord was well imaged in its entire craniocaudal extent, comparable to CT myelography. The bony septum, when it contained a marrow cavity, was also seen well. In two patients, dural ectasia and low position of the spinal cord with and without associated lipoma were clearly imaged. MR imaging demonstrated associated syringohydromyelia in one patient that was not detected by other radiologic studies. This preliminary experience with MR imaging of diastematomyelia suggests that once the bony details of the abnormality are defined, MR imaging can delineate the presence and extent of the divided spinal cord as well as its associated abnormalities adequately, obviating other studies.

Acute fractures and dislocations of the cervical spine. An analysis of three hundred hospitalized patients and review of the literature.

Of 300 patients who were hospitalized for acute cervical injuries, 216 lived, fifty-one died within four months of injury, and thirty-three were lost to follow-up. The important findings in a retrospective review were that laminectomy resulted in a high mortality rate and loss of motor function and that steroids did not improve neural recovery in quadriplegics and their use was associated with gastrointestinal hemorrhage. Closed or open reduction followed by posterior fusion for subluxations or dislocations, and anterior decompression and fusion for vertebral compression fractures, offered the best chance for recovery of neural function and restoration of stability. Massive epidural hemorrhage was found only in patients with ankylosing spondylitis.

One-stage anterior cervical decompression and posterior stabilization with circumferential arthrodesis. A study of twenty-four patients who had a traumatic or a neoplastic lesion.

Twenty-four patients had a combined anterior cervical decompression and posterior stabilization with circumferential spinal arthrodesis for treatment of either a tumor or an injury. The indication for operation was a fixed kyphosis and an incomplete neurological deficit or cervical instability. All but two patients had substantial improvement, having regained strength or had a reduction in the deformity, or both. The two exceptional patients, both of whom were quadriparetic, had no change. The operation is formidable and requires an average of 6.9 hours of general anesthesia; however, its use is justified in patients who have the appropriate indications.

Complications following Harrington instrumentation for fractures of the thoracolumbar spine.

The purposes of using Harrington instrumentation for the treatment of thoracolumbar fractures are to reduce the fracture, decompress the spinal canal, create stability at the fracture site, and shorten the hospitalization period. However, technical problems or the injudicious use of Harrington-instrumentation systems can also complicate the management of these fractures. We have studied forty patients (forty-five Harrington-instrumentation stabilization procedures) who had significant complications. Twenty-six of the thirty patients who were followed for more than two years required additional spinal reconstructive surgical procedures. Five patients had neurological deterioration (one died), nine patients had an inadequate reduction of translational displacement of a vertebral fracture, sixteen patients had dislodgment or disengagement of the Harrington components with resultant loss of fixation, six patients had a deep wound infection, three patients had a complete wound dehiscence with exposure of metal, and sixteen patients had persistent unrecognized neural compression. Several factors were associated with these failures of Harrington instrumentation: translational (flexion-rotation) injuries of the osteoligamentous middle column; failure to obtain either myelographic or computed tomographic studies, or both, postoperatively; failure to identify persistent neural compression; wound dehiscence; the use of distraction rods for high thoracic kyphosis; and instrumentation across the lumbosacral joint.

Occipitocervical fusion. Indications, technique, and long-term results in thirteen patients.

Thirteen patients who underwent occipitocervical fusion that was performed using the same operative technique were followed for an average of 3.6 years (range, two to seven years). The indications for surgery were occipitocervical instability, neurological deficit, or intractable pain that was not responsive to non-operative treatment. Of the thirteen patients, eight had rheumatoid arthritis, two had atlanto-axial osteomyelitis, and one patient each had trauma, ankylosing spondylitis, and atlanto-occipital osteoarthritis. At follow-up, all of the thirteen patients had a solid arthrodesis and relief of severe pain in the neck. Of the ten patients who had had myelopathy preoperatively, all improved, but of four patients who had been unable to walk preoperatively because of severe motor involvement, only one was considered to be able to walk. Of the thirteen patients, ten had a satisfactory result. Occipitocervical arthrodesis using iliac grafts and the wiring technique that is described herein achieves immediate stable fixation, allowing early mobilization with a successful arthrodesis, and it should be undertaken before severe myelopathy occurs in patients who have instability of the cervical spine. The operation may optimize the patient's chances of neurological recovery.

Anterior excision of herniated thoracic discs.

Twenty-two herniations of a thoracic disc in nineteen patients were surgically excised between 1972 and 1984. An anterior transthoracic decompression or a costotransversectomy was used for all discectomies. Pain and paraparesis were the most common symptoms preoperatively. The average length of follow-up was forty-eight months. Sixteen patients had an excellent or a good result, one had a fair result, and two had a poor result. Twelve of the fourteen patients who had had motor weakness preoperatively had varying degrees of improvement in motor function postoperatively. Pain was relieved in ten patients, reduced in eight, and unchanged in one. Anterior decompression of herniated thoracic discs yielded gratifying results, but the procedure is associated with some risk of damage to the spinal cord. It therefore requires meticulous preoperative planning and careful surgical technique.

Mediastinal widening associated with fractures of the upper thoracic spine.

Widening of the mediastinum, when seen on radiographs of the chest in victims of trauma, is usually attributed to injury to the aorta. An aortic injury, when not lethal, often causes paraparesis or paraplegia due to ischemia of the spinal cord. A fracture of the upper thoracic spine can produce similar clinical and radiographic findings. The cases of three patients who had those findings are presented; in all three, the differential diagnosis between the vascular and skeletal injuries was difficult. Fracture of the thoracic spine should be included in the differential diagnosis whenever mediastinal widening is seen on radiographs.

Cervical kyphosis and myelopathy. Treatment by anterior corpectomy and strut-grafting.

Between 1976 and 1984, fourteen patients who had severe cervical kyphosis and myelopathy were treated with anterior decompression and arthrodesis. Eight had had spondylosis; five, a traumatic injury; and one, a benign intradural tumor. In eight of the fourteen patients, the severe kyphosis and myelopathy had developed after a laminectomy of three, four, or five cervical vertebrae. The laminectomy had been done for the treatment of spondylosis in five patients, of a traumatic lesion in two, and of a tumor in one. Considering all fourteen patients, an average of 2.25 vertebral bodies was removed from each, and the average extent of the subsequent fusion was 3.25 levels. Eight patients (six of whom had spondylosis; one, a traumatic lesion; and one, a tumor) were treated with a fibular graft that spanned an average of 4.10 levels, and six patients (four of whom had a traumatic lesion and two, spondylosis) were treated with an iliac graft that spanned an average of 2.70 levels. Of the five patients who had a traumatic lesion, four were treated with anterior decompression and arthrodesis, combined with posterior arthrodesis that was performed during the same period of anesthesia. In three patients, the anterior graft dislodged during the immediate postoperative period. Two of the three patients had posterior instability due to a prior laminectomy, and in the third the graft dislodged because of technical difficulties. Two of these grafts were revised to restore stability. At the latest follow-up, twelve of the fourteen fusions were solid. In the other two patients, who died six and ten months postoperatively, the fusion had been solid, as shown by radiographs, before the time of death. The average amount of correction of the kyphotic deformities was 32 degrees, a reduction from an average of 45 degrees to an average of 13 degrees. All but one patient had some recovery of neural function; nine had complete and four, partial recovery. The remaining patient had relief of pain, but he continued to be completely quadriplegic although he had some sensory sparing. Of the four patients who had been unable to walk preoperatively, three were able to walk postoperatively. No patient lost neural function after the anterior decompression and arthrodesis. We concluded that, in the presence of severe cervical kyphosis and myelopathy, adequate anterior decompression of the spinal cord, correction of the kyphosis, and anterior arthrodesis using a strut graft can yield excellent results without undue risk.

Spondylolisthesis treated by a single-stage operation combining decompression with in situ posterolateral and anterior fusion. An analysis of eleven patients who had long-term follow-up.

We treated eleven skeletally mature patients who had a high-grade lumbosacral spondylolisthesis by a single-stage operation that involved posterior spinal decompression, posterolateral arthrodesis with autogenous iliac-crest graft, and anterior arthrodesis with a fibular graft inserted from the posterior approach. We did not attempt to correct the deformity. Preoperatively, all but one patient had severe pain in the back and lower limb; the exceptional patient had severe pain only in the low back. All but one patient had sensory deficits and objective motor impairment before the operation, and five had cauda equina syndrome. Six patients had had a previous operation that had failed. The duration of follow-up ranged from two to twelve years. A solid fusion was obtained in all patients, and all had major or complete neurological recovery. The results did not deteriorate with time.

Anterior decompression and arthrodesis of the cervical spine: long-term motor improvement. Part II--Improvement in complete traumatic quadriplegia.

Fifty-one patients who had complete motor quadriplegia secondary to a fracture or dislocation of the cervical spine were managed by anterior cervical decompression and arthrodesis with iliac bone grafts between 1973 and 1983. In all patients, myelography demonstrated that displaced fragments of bone and disc were compressing the anterior aspect of the spinal cord. Decompression was performed in an attempt to gain further improvement of the motor-roots in the upper extremities and thereby to improve the ability of the patients to care for themselves. The average interval from the injury to the decompression was fifteen months (range, one month to eight years). Two patients died within two months after the operation, one had a respiratory arrest that resulted in brain damage one day after the operation, and two died from cardiovascular disease more than one year after the operation. The remaining forty-six patients were followed for an average of five years (range, two to thirteen years). Neurological improvement of at least two new functional motor-root levels was documented in seven patients and of one level, in eighteen. Increased motor strength by two or three grades was seen in six patients. Noteworthy motor improvement did not occur in the remaining twenty patients. The mean modified Barthel index (used to measure improvement in the ability to perform activities of daily living) increased from 17 to 33 (of a possible 100) points. Functionally important improvement of the caudad part of the cord occurred in only one patient. In one patient, neural injury, with loss of one motor-root level, occurred, with only partial improvement. At the latest follow-up examination, the result was poor in nine of eleven patients who had decompression eighteen months or more after the injury. Only two of these patients had any improvement in the Barthel index, and then of only 5 points each. The result also was poor in the five patients who were more than fifty-three years old; two had no improvement in the Barthel index, one improved by 5 points, and two died.

Anterior decompression and arthrodesis of the cervical spine: long-term motor improvement. Part I--Improvement in incomplete traumatic quadriparesis.

Between 1973 and 1983, fifty-eight patients who had an incomplete spinal-cord injury secondary to a fracture or dislocation of the cervical spine were managed by anterior cervical decompression and arthrodesis with iliac bone grafts. In all patients, myelography showed that displaced fragments of bone or disc were compressing the anterior aspect of the spinal cord. Anterior decompression was performed in an attempt to improve function in the upper and lower extremities. The average interval from the injury to the decompression was thirteen months (range, one month to nine years). Two patients died of cardiopulmonary disease within two months after the operation, and one patient died eighteen months after the operation. The remaining fifty-five patients were followed for an average of six years (range, two to seventeen years). Twenty-nine patients became functional ambulators after the operation. An additional six patients who could walk before the operation had improvement in the ability to walk. Noteworthy improvement in motor-root function in the upper extremities was seen in thirty-nine patients. Only nine patients had no signs of improvement of motor function. Improvement was less in the patients in whom operative decompression had been done more than twelve months after the injury. The patients who had an extension injury to a spondylotic spine were older, and fewer of them had improvement. No patient lost neurological function as a result of the operation. Anterior decompression and arthrodesis, even when performed late after the injury, can improve neurological function in both the upper and lower extremities in many patients who have incomplete quadriplegia due to a fracture or dislocation of the cervical spine.

The results of treatment of acute injuries of the upper thoracic spine with paralysis.

We studied the cases of 218 patients who had had trauma to the upper region of the thoracic spine (first to tenth thoracic vertebra) that resulted in paralysis. Of these, 184 patients had a complete and thirty-four had an incomplete lesion of the spinal cord. Of the 184 patients who had a complete lesion of the spinal cord, none of the 149 patients who were followed for two to fifteen years recovered any significant neurological function, regardless of the type of operative or non-operative treatment. Thirty of the patients with an incomplete lesion of the spinal cord were followed for two to twenty years. Three others were lost to follow-up, and one died with ascending necrosis of the spinal cord and pulmonary failure three months after the laminectomy. Of the remaining thirty patients, twenty-three had an anterior cord syndrome, four had a central cord syndrome, and three had a Brown-Séquard syndrome. Five patients with an incomplete lesion were treated without surgery. Three of these patients recovered the ability to walk while two recovered some function but were still not able to walk. Seventeen patients with an incomplete lesion of the spinal cord were treated by laminectomy. Of these patients, two also had cord-cooling, two had a posterior fusion, and one had cord-cooling, spine fusion, and posterior instrumentation with Harrington rods. Five patients recovered partial ability to walk with braces, four did not, and eight lost neurological function or became completely paraplegic after surgery and did not recover. Eight patients with an incomplete lesion of the spinal cord were treated with anterior transthoracic decompression and fusion, three of whom had had a previous laminectomy that had not improved their status. Five patients recovered the ability to walk without aids, two walked with braces, and one recovered some motor function but was not able to walk. From this study, we concluded that laminectomy is contraindicated for incomplete lesions of the upper region of the thoracic spinal cord and that anterior transthoracic decompression and fusion offers the best chance of recovery of neurological function.

Osteoarthrosis of the atlanto-axial joints. Long-term follow-up after treatment with arthrodesis.

We evaluated the results for fifteen patients in whom symptomatic osteoarthrosis of the atlanto-axial joints had been treated with an atlanto-axial or occipitocervical arthrodesis between 1973 and 1990. Thirteen patients had long-term follow-up (average duration, seven years and two months; range, four years and two months to nineteen years and two months). The two remaining patients had died: one, four days postoperatively, from a cardiopulmonary arrest, and the other, one year and eight months postoperatively, from complications related to bladder cancer. Preoperatively, all fifteen patients reported pain in the occipitocervical region that increased with any attempt at rotation of the neck and was unresponsive to immobilization with a collar and to analgesics. The average duration of the symptoms before the arthrodesis was three years. One patient had acute quadriparesis. All patients had radiographic evidence of oesteoarthrosis involving the lateral atlanto-axial articulations. Four patients had atlanto-axial instability with an average of five millimeters (range, three to ten millimeters) of motion at the anterior atlanto-odontoid interval. Six patients had an associated spontaneous subaxial fusion, which was secondary to osteoarthrosis in five; three of the five also had atlanto-axial instability. Fourteen patients were managed with a posterior arthrodesis and one, with an anterior transoral arthrodesis. The procedures were performed to relieve pain, to stabilize the atlanto-axial joints, and to restore neurological function. Of the fourteen patients who were followed, thirteen had a solid fusion and one had a stable pseudarthrosis. The patient who had quadriparesis recovered. At the latest follow-up evaluation, thirteen patients had an excellent result and one had a fair result as determined with use of a modification of the criteria of Robinson et al. There were no poor results. Atlanto-axial arthrodesis can effectively relieve occipitocervical pain and correct atlanto-axial instability secondary to osteoarthrosis.

Anterior decompression of traumatic thoracolumbar fractures with incomplete neurological deficit using a retroperitoneal approach.

Between 1973 and 1981, seventy patients with a spinal cord injury secondary to a thoracolumbar fracture were treated by anterior spinal-canal decompression through a retroperitoneal approach. All of these patients had an incomplete neurological deficit caused by retropulsed vertebral-body fragments and intervertebral disc material in the spinal canal. Forty-eight patients have been followed for an average of 3.4 years (range, two to 8.6 years). Either computed tomography or lateral tomography, or both, was performed after surgery on these forty-eight patients, and confirmed the successful removal of the cause of compression in all of them. No patient lost further cord or cauda equina function after the anterior decompression. Thirty-seven of the forty-two patients who had a motor deficit improved by at least one class in motor strength. Fourteen of the thirty patients whose quadriceps and hamstrings were too weak to permit walking regained full independent walking ability. Twelve of the thirty-two patients who had a conus medullaris injury demonstrated neurogenic bowel and bladder recovery. The degree of neurological recovery of spinal cord injury after anterior spinal decompression of thoracolumbar fractures appears more favorable than after other, previously reported techniques that do not decompress the spinal canal.

The use of epidural morphine to decrease postoperative pain in patients undergoing lumbar laminectomy.

In a double-blind study of fifty patients that was done to test the ability of epidural morphine to decrease the discomfort after lumbar laminectomy, we found that during the first postoperative day the patient's pain was lessened appreciably but that the total dose of morphine received postoperatively was not diminished. We used a three-milligram dose of epidural morphine, and there were no problems with respiratory depression. The addition of epinephrine to the morphine solution had no beneficial effect.

Charcot's arthropathy of the spine following spinal cord injury. A report of five cases.

Between 1955 and 1968, we treated five patients who had post-traumatic Charcot's arthropathy of the spine. All of the patients had a neurologically complete spinal-cord lesion. A laminectomy was implicated as the cause in four of the five patients. Spinal arthrodesis was used successfully to correct the deformity, stabilize the spine, restore sitting balance, and prevent complications resulting from neuropathic arthropathy of the vertebrae following complete spinal-cord injury.

Experimental lumbar spinal stenosis. Analysis of the cortical evoked potentials, microvasculature, and histopathology.

An animal model of lumbar spinal stenosis was developed in which the pathophysiology of this condition could be examined. Four experimental groups, each containing six dogs, were studied. One group had a laminectomy of the sixth and seventh lumbar vertebrae only; these animals served as controls. In the three other groups, a laminectomy was performed and the cauda equina was constricted by 25, 50, or 75 per cent to produce chronic compression. Cortical evoked potentials were recorded preoperatively, immediately after constriction, and at one, two, and three months postoperatively. Daily neurological examinations were carried out, and the neurological deficits were graded using the Tarlov system. After three months of constriction, the cauda equina of three dogs in each group was examined histologically, and the vascular circulation was examined by latex and India-ink injection with a modification of the Spalteholz technique. The animals in the control group showed no neurological abnormalities, no changes in cortical evoked potentials, normal microvascularity, and no histopathological changes in the nerve roots or the spinal cord. The dogs in which the cauda equina had been constricted 25 per cent had no neurological deficits, mild changes in cortical evoked potentials, slight histological changes, and venous congestion of the root and dorsal root ganglion of the seventh lumbar nerve. The dogs in which the cauda equina had been constricted 50 per cent had mild initial motor weakness, major changes in cortical evoked potentials, edema and loss of myelin in the root of the seventh lumbar nerve, and moderate or severe venous congestion of the root and dorsal root ganglion of the seventh lumbar nerve. The dogs in which the cauda equina had been constricted 75 per cent had significant weakness, paralysis of the tail, and urinary incontinence; two dogs recovered by the third month, but all had neurogenic claudication for three months. All six dogs had dramatic changes in cortical evoked potentials and had complete nerve-root atrophy at the level of the constriction. There was blockage of axoplasmic flow and wallerian degeneration of the motor nerve roots distal to the constriction and of the sensory roots proximal to the constriction, as well as degeneration of the posterior column. Severe arterial narrowing at the level of the constriction and venous congestion of the roots and dorsal root ganglia of the seventh lumbar and first sacral nerves were also present. Cortical evoked potentials revealed neurological abnormalities before the appearance of neurological signs and symptoms.(ABSTRACT TRUNCATED AT 400 WORDS)

Upper-airway obstruction after multilevel cervical corpectomy for myelopathy.

We reviewed the cases of seven patients who had had obstruction of the upper airway immediately after an anterior procedure on the cervical spine and had required reintubation. All patients had had moderate or severe myelopathy preoperatively (average, 3.6 on the Nurick scale), and all had had a multilevel anterior cervical corpectomy for decompression followed by arthrodesis. The early compromise of the upper airway was believed to be due to edema rather than to the formation of a hematoma. Five patients had no sequelae, but two died of complications related to the obstruction. Risk factors common to these patients included moderate or severe myelopathy and multilevel corpectomy. Six patients had a history of heavy smoking and one, of asthma. We believe that extra caution should be used in the postoperative management of the airways when multilevel corpectomy is performed in patients who have these pre-existing conditions.

Sagittal measurements of the cervical spine in subaxial fractures and dislocations. An analysis of two hundred and eighty-eight patients with and without neurological deficits.

We analyzed three factors involved in fractures and dislocations of the cervical spine and their relation to the degree of injury of the spinal cord. The three factors were the space available for the spinal cord at the level of the injury, the sagittal diameter of the spinal canal at the uninjured levels, and the Pavlov ratio at the uninjured levels. Of the 288 patients analyzed, eighty-three had a complete injury of the spinal cord, ninety-two had an incomplete injury of the spinal cord, thirty had an isolated nerve-root injury, and eighty-three had no neurological deficit. The mean space available for the spinal cord at the level of the injury was 10.5 millimeters for the patients who had a complete injury of the spinal cord, 13.1 millimeters for those who had an incomplete injury of the spinal cord, 15.9 millimeters for those who had an isolated nerve-root injury, and 16.7 millimeters for those who had no neurological deficit. The difference between the groups was significant (p < 0.001) except for the difference between the patients who had an isolated nerve-root injury and those who had no neurological deficit. The mean sagittal diameter of the canal at the uninjured levels was 16.1 millimeters for the patients who had a complete injury of the spinal cord, 16.1 millimeters for those who had an incomplete injury of the spinal cord, 17.9 millimeters for those who had an isolated nerve-root injury, and 18.1 millimeters for those who had no neurological deficit. The difference between the patients who had a complete injury of the spinal cord and those who had an incomplete injury of the spinal cord and that between the patients who had an isolated nerve-root injury and those who had no neurological deficit were not significant (p > 0.05). However, the patients who had a complete injury of the spinal cord and those who had an incomplete injury of the spinal cord were significantly different from the patients who had an isolated nerve-root injury and those who had no neurological deficit (p < 0.001). The mean Pavlov ratio at the uninjured levels was 0.82 for the patients who had a complete injury of the spinal cord, 0.84 for those who had an incomplete injury of the spinal cord, 0.96 for those who had an isolated nerve-root injury, and 0.96 for those who had no neurological deficit.(ABSTRACT TRUNCATED AT 400 WORDS)