Eric M. Horn, Nicholas Theodore, Neil R. Crawford, Nicholas C. Bambakidis and Volker K. H. Sonntag
Lateral mass screws are traditionally used to fixate the subaxial cervical spine, while pedicle screws are used in the thoracic spine. Lateral mass fixation at C-7 is challenging due to thin facets, and placing pedicle screws is difficult due to the narrow pedicles. The authors describe their clinical experience with a novel technique for transfacet screw placement for fixation at C-7.
A retrospective chart review was undertaken in all patients who underwent transfacet screw placement at C-7. The technique of screw insertion was the same for each patient. Polyaxial screws between 8- and 10-mm-long were used in each case and placed through the facet from a perpendicular orientation. Postoperative radiography and clinical follow-up were analyzed for aberrant screw placement or construct failure.
Ten patients underwent C-7 transfacet screw placement between June 2006 and March 2007. In all but 1 patient screws were placed bilaterally, and the construct lengths ranged from C-3 to T-5. One patient with a unilateral screw had a prior facet fracture that precluded bilateral screw placement. There were no intraoperative complications or screw failures in these patients. After an average of 6 months of follow-up there were no hardware failures, and all patients showed excellent alignment.
The authors present the first clinical demonstration of a novel technique of posterior transfacet screw placement at C-7. These results provide evidence that this technique is safe to perform and adds stability to cervicothoracic fixation.
Cervical magnetic resonance imaging abnormalities not predictive of cervical spine instability in traumatically injured patients
Invited submission from the Joint Section Meeting on Disorders of the Spine and Peripheral Nerves, March 2004
Eric M. Horn, Gregory P. Lekovic, Iman Feiz-Erfan, Volker K. H. Sonntag and Nicholas Theodore
Object. Identifying instability of the cervical spine can be difficult in traumatically injured patients. The goal of this study was to determine whether cervical abnormalities demonstrated on magnetic resonance (MR) imaging are predictive of spinal instability.
Methods. Data in all patients admitted through the Level I trauma service at the authors' institution who had undergone cervical MR imaging were retrospectively reviewed. The reasons for MR imaging screening were neurological deficit, fracture, neck pain, and indeterminate clinical examination (for example, coma). Abnormal soft-tissue (prevertebral or paraspinal) findings on MR imaging were correlated with those revealed on computerized tomography (CT) scanning and plain and dynamic radiography to determine the presence/absence of cervical instability.
Of 6328 patients admitted through the trauma service, 314 underwent MR imaging of the cervical spine. Of 166 patients in whom CT scanning or radiography demonstrated normal findings, 70 had undergone MR imaging that revealed abnormal findings. Of these 70 patients, 23 underwent dynamic imaging, the findings of which were normal. In each case of cervical instability (65 patients) CT, radiographic, and MR imaging studies demonstrated abnormalities. Furthermore, there were 143 patients with abnormal CT or radiographic study findings, in 13 of whom MR imaging revealed normal findings. Six of the latter underwent dynamic testing, which demonstrated normal results.
Conclusions. Magnetic resonance imaging is sensitive to soft-tissue injuries of the cervical spine. When CT scanning and radiography detect no fractures or signs of instability, MR imaging does not help in determining cervical stability and may lead to unnecessary testing when not otherwise indicated.
Report of three cases
Eric M. Horn, Ruth E. Bristol, Iman Feiz-Erfan, Elisa J. Beres, Nicholas C. Bambakidis and Nicholas Theodore
✓Pseudomeningoceles rarely develop after cervical trauma; in all reported cases the lesions have extended outside the spinal canal.
The authors report the first known cases of anterior cervical pseudomeningoceles contained entirely within the spinal canal and causing cord compression and neurological injury. The authors retrospectively reviewed the cases of three patients with traumatic cervical spine injuries and concomitant compressive anterior pseudomeningoceles. The lesion was recognized in the first case when the patient’s neurological status declined after he sustained a severe atlantoaxial injury; the pseudomeningocele was identified intraoperatively and decompressed. After the decompressive surgery, the patient’s severe tetraparesis partially resolved. In the other two patients diagnoses of similar pseudomeningoceles were established by magnetic resonance imaging. Both patients were treated conservatively, and their mild to moderate hemiparesis due to the pseudomeningocele-induced compression abated.
The high incidence of anterior cervical pseudomeningoceles seen at the authors’ institution within a relatively brief period suggests that this lesion is not rare. The authors believe that it is important to recognize the compressive nature of these lesions and their potential to cause devastating neurological injury.
Report of two cases and review of the literature
Gregory P. Lekovic, Nitin R. Mariwalla, Eric M. Horn, Steven Chang, Harold L. Rekate and Nicholas Theodore
✓ Because skeletal dysplasias are primary disorders of bone, they have not been commonly understood as neurosurgical diseases. Nevertheless, neurosurgical complications are commonly encountered in many cases of dysplasia syndromes. The authors present two cases of skeletal dysplasia that caused overt instability of the cervical spine. One patient with a diagnosis of Gorham disease of the cervical spine was treated with prolonged fixation in a halo brace after an initial attempt at instrumentation with a posterior occiput–C4 fusion. The other patient, who at birth was identified to have camptomelic dysplasia, has been treated conservatively from the outset. Although these two patients presented with different disorders—in one patient adequate mature bone never formed and in the other patient progressive bone loss became apparent after a seemingly normal initial development—these cases demonstrate unequivocally that surgical options for fusion are ultimately limited by the quality of the underlying bone. In patients in whom the bone itself is inadequate for use as a substrate for fusion, there are currently limited treatment options. Future improvements in our understanding of chondrogenesis and ossification may lead to the design of superior methods of encouraging fusion in these patients; however, at the present time, long-term maintenance in a halo brace may, in fact, be the only treatment.
Eric M. Horn, Nicholas Theodore, Rachid Assina, Robert F. Spetzler, Volker K. H. Sonntag and Mark C. Preul
Venous stasis and intrathecal hypertension are believed to play a significant role in the hypoperfusion present in the spinal cord following injury. Lowering the intrathecal pressure via cerebrospinal fluid (CSF) drainage has been effective in treating spinal cord ischemia during aorta surgery. The purpose of the present study was to determine whether CSF drainage increases spinal cord perfusion and improves outcome after spinal injury in an animal model.
Anesthetized adult rabbits were subjected to a severe contusion spinal cord injury (SCI). Cerebrospinal fluid was then drained via a catheter to lower the intrathecal pressure by 10 mm Hg. Tissue perfusion was assessed at the site of injury, and values obtained before and after CSF drainage were compared. Two other cohorts of animals were subjected to SCI: 1 group subsequently underwent CSF drainage and the other did not. Results of histological analysis, motor evoked potential and motor function testing were compared between the 2 cohorts at 4 weeks postinjury.
Cerebrospinal fluid drainage led to no significant improvement in spinal cord tissue perfusion. Four weeks after injury, the animals that underwent CSF drainage demonstrated significantly smaller areas of tissue damage at the injury site. There were no differences in motor evoked potentials or motor score outcomes at 4 weeks postinjury.
Cerebrospinal fluid drainage effectively lowers intrathecal pressure and decreases the amount of tissue damage in an animal model of spinal cord injury. Further studies are needed to determine whether different draining regimens can improve motor or electrophysiological outcomes.
Eric M. Horn, Iman Feiz-Erfan, Gregory P. Lekovic, Curtis A. Dickman, Volker K. H. Sonntag and Nicholas Theodore
Although rare, traumatic occipitoatlantal dislocation (OAD) injuries are associated with a high mortality rate. The authors evaluated the imaging and clinical factors that determined treatment and were predictive of outcomes, respectively, in survivors of this injury.
The medical records and imaging studies obtained in 33 patients with OAD were reviewed retrospectively. Clinical factors that predicted outcomes, especially neurological injury at presentation and imaging findings, were evaluated.
The most sensitive method for the diagnosis of OAD was the measurement of basion axial–basion dens interval on computed tomography (CT) scanning. Five patients with severe traumatic brain injuries (TBIs) were not treated and subsequently died. Of the 28 patients in whom treatment was performed, 23 underwent fusion and five were fitted with an external orthosis. Abnormal findings of the occipitoatlantal ligaments on magnetic resonance (MR) imaging, associated with no or questionable abnormalities on CT scanning, provided the rationale for nonoperative treatment. Of the 28 patients treated for their injuries, perioperative death occurred in five, three of whom had presented with severe neurological injuries. The mortality rate was highest in patients with a TBI at presentation. The mortality rate was lower in patients presenting with a spinal cord injury, but in this group there was a higher rate of persistent neurological deficits.
The spines in patients with CT-documented OAD are most likely unstable and need surgical fixation. In patients for whom CT findings are normal and MR imaging findings suggest marginal abnormalities, nonoperative treatment should be considered. The best predictors of outcome were severe brain or upper cervical injuries at initial presentation.
Eric M. Horn, Jonathan S. Hott, Randall W. Porter, Nicholas Theodore, Stephen M. Papadopoulos and Volker K. H. Sonntag
✓ Atlantoaxial stabilization has evolved from simple posterior wiring to transarticular screw fixation. In some patients, however, the course of the vertebral artery (VA) through the axis varies, and therefore transarticular screw placement is not always feasible. For these patients, the authors have developed a novel method of atlantoaxial stabilization that does not require axial screws. In this paper, they describe the use of this technique in the first 10 cases.
Ten consecutive patients underwent the combined C1–3 lateral mass–sublaminar axis cable fixation technique. The mean age of the patients was 62.6 years (range 23–84 years). There were six men and four women. Eight patients were treated after traumatic atlantoaxial instability developed (four had remote trauma and previous nonunion), whereas in the other two atlantoaxial instability was caused by arthritic degeneration. All had VA anatomy unsuitable to traditional transarticular screw fixation.
There were no intraoperative complications in any of the patients. Postoperative computed tomography studies demonstrated excellent screw positioning in each patient. Nine patients were treated postoperatively with the aid of a rigid cervical orthosis. The remaining patient was treated using a halo fixation device. One patient died of respiratory failure 2 months after surgery. Follow-up data (mean follow-up duration 13.1 months) were available for seven of the remaining nine patients and demonstrated a stable construct with fusion in each patient.
The authors present an effective alternative method in which C1–3 lateral mass screw fixation is used to treat patients with unfavorable anatomy for atlantoaxial transarticular screw fixation. In this series of 10 patients, the method was a safe and effective way to provide stabilization in these anatomically difficult patients.
Eric M. Horn, Nicholas Theodore, Iman Feiz-Erfan, Gregory P. Lekovic, Curtis A. Dickman and Volker K. H. Sonntag
The risk factors of halo fixation in elderly patients have never been analyzed. The authors therefore retrospectively reviewed data obtained in the treatment of such cases.
A discharge database was searched for patients 70 years of age or older who had undergone placement of a halo device. In a search of cases managed between April 1999 and February 2005, data pertaining to 53 patients (mean age 79.9 years [range 70–97 years]) met these criteria. Forty-one patients were treated for traumatic injuries. Ten patients had deficits ranging from radiculopathy to quadriparesis, and 43 had no neurological deficit. Adequate follow-up material was available in 42 patients (mean treatment duration 91 days). Halo immobilization was the only treatment in 21 patients, and adjunctive surgical fixation was undertaken in the other 21 patients. There were 31 complications in 22 patients: respiratory distress in four patients, dysphagia in six, and pin-related complications in 10. Eight patients died; in two of these cases, the cause of death was clearly unrelated to the halo brace. The other six patients died of respiratory failure and cardiovascular collapse (perioperative mortality rate 14%). Three patients who died had sustained acute trauma and three had undergone surgical stabilization.
External halo fixation can be used safely to treat cervical instability in elderly patients. The high complication rate in this population may reflect the significant incidence of underlying disease processes.
Eric M. Horn, Phillip M. Reyes, Seungwon Baek, Mehmet Senoglu, Nicholas Theodore, Volker K. H. Sonntag and Neil R. Crawford
The small diameter of the pedicle can make C-7 pedicle screw insertion dangerous. Although transfacet screws have been studied biomechanically when used in pinning joints, they have not been well studied when used as part of a C7–T1 screw/rod construct. The authors therefore compared C7–T1 fixation using a C-7 transfacet screw/T-1 pedicle screw construct with a construct composed of pedicle screws at both levels.
Each rigid posterior screw/rod construct was placed in 7 human cadaveric C6–T2 specimens (14 total). Specimens were tested in normal condition, after 2-column instability, and once fixated. Nondestructive, nonconstraining pure moments (maximum 1.5 Nm) were applied to induce flexion, extension, lateral bending, and axial rotation while recording 3D motion optoelectronically. The entire construct was then loaded to failure by dorsal linear force.
There was no significant difference in angular range of motion between the 2 instrumented groups during any loading mode (p > 0.11, nonpaired t-tests). Both constructs reduced motion to < 2° in any direction and allowed significantly less motion than in the normal condition. The C-7 facet screw/T-1 pedicle screw construct allowed a small but significantly greater lax zone than the pedicle screw/rod construct during lateral bending, and it failed under significantly less load than the pedicle screw/rod construct (p < 0.001).
When C-7 transfacet screws are connected to T-1 pedicle screws, they provide equivalent stability of constructs formed by pedicle screws at both levels. Although less resistant to failure, the transfacet screw construct should be a viable alternative in patients with healthy bone.