Circumferential Cervical Spinal Fusion

Published on 13/03/2015 by admin

Last modified 22/04/2025

Print this page

This article have been viewed 2888 times

Chapter 158 Circumferential Cervical Spinal Fusion

Recent advances in complex spinal surgery have helped foster the increased number of cervical spine procedures that are using a bidirectional approach. This is done for decompression of the neural elements, biomechanical stresses, and to influence the fusion ability of the patient. Although the ability of most surgeons allows them to use this technique to achieve surgical goals, it is also not without potential drawbacks. These issues can include surgical time, morbidity to the patient, and substantial increased cost. The purpose of this chapter is to provide details of patients who may benefit from a circumferential approach, followed by discussion of proper technique and complication awareness/avoidance, and a review of the outcomes.

The Rationale for Circumferential Surgery

It is perhaps useful to first review the goals of surgery on the cervical spine, so as to best put each item in proper context and how it may influence the choice to consider circumferential surgery for a given patient. These include decompression of the neural elements, restoration of spinal alignment, spinal stability, proper environment for fusion, and the health of the patient.

Certainly one of the most obvious reasons for choosing bidirectional surgery of the cervical spine would be that the patient presents with obvious compression of both the ventral and dorsal surfaces of the spinal cord (Fig. 158-1). If the patient is not overly frail, then one can make a rationale for this approach. There is some degree of inherent common sense to directly remove the compressive structure of the spinal cord. One variation might be surgery to treat ossification of the posterior longitudinal ligament (OPLL).¹^–⁴ Patients with dramatic anterior compression are commonly treated by posterior decompression in order to avoid the significant pitfalls of anterior surgery.

FIGURE 158-1 T2-weighted MRI demonstrating both anterior and posterior spinal cord compression at the C4–C5 and C5–C6 levels from a combination of disc herniations and spondylitic stenosis. Abnormal “high” signal changes are also noted, indicating spinal cord edema or myelomalacia.

Restoration of spinal alignment has received an increasing degree of attention recently as a potential crucial factor in the outcome of patients with cervical spondylitic myelopathy (CSM).⁵^–⁸ Most would agree that circumferential surgery is indicated for patients that present in frank kyphosis (Fig. 158-2). This may result from numerous factors including posterior migration of the cord, vascular perfusion, and longitudinal stresses on the cord. If a patient is left in a kyphotic position, then draping and potential anterior compression of the spinal cord are certainly possible. The classic example of this would be postlaminectomy kyphosis. Nottmeier et al. recently published results of cervical kyphotic deformity correction using a anterior-posterior reconstruction. Their patient population had a mean preoperative kyphosis of 18 degrees. They were able to achieve a mean correction of sagittal angle of 22 degrees. This resulted in a mean postoperative sagittal angle of 4 degrees. There was no reported loss of deformity correction during the follow-up period and they reported a fusion rate of 97.5%.⁸

FIGURE 158-2 T2-weighted MRI in a patient with a previous fusion of C5–C7 with a pronounced anterolisthesis and kyphosis of C4–C5.

Several studies have looked at the potential importance of postoperative posterior spinal cord migration. There is a concern that residual kyphosis may allow the spinal cord to be draped across the vertebral column, which may contribute to compromise of the spinal cord via vascular compromise or longitudinal stress. Lee et al. attempted to quantitatively assess preoperative factors that would predict this postoperative movement of the cord. Besides the commonly stated importance of preoperative sagittal balance, they also found significance in the relative degree of stenosis directly above the “laminectomized” level to be particularly important.⁹ However, the clinical importance of cord shift remains uncertain. Hatta et al. found no clinical correlation in myelopathy outcomes to cord migration.¹⁰

There has also been some attention given to the potential effects of “longitudinal stress” to the spinal cord by Uchida et al. It is interesting to consider the potential similarities of this to the physiologic derangement that has been documented to occur with tethering of the spinal cord.^11,¹²

History

Bidirectional surgery for the cervical spine was originally reserved for cases of clear violation of stability of both the anterior and posterior elements of the cervical spine. Most cases dealt with either trauma or neoplastic disease. Instrumentation of the spine was limited to wiring techniques (Figs. 158-3, 158-4, and 158-5). Cases frequently required staging, and patients commonly required prolonged immobilization.

FIGURE 158-3 Posterior stainless steel wires with a solid posterior fusion.

FIGURE 158-4 Patient with preoperative kyphosis corrected with anterior fibular strut fusion with posterior wiring technique. A triple-wire technique is used.

FIGURE 158-5 A, A fixed kyphosis in a patient with neck and C6 radicular pain with chin-on-chest deformity. B, Corrective surgery with an anterior discectomy and allograft reconstruction and posterior triple-wire fusion.

The evolution of surgical hardware has taken us from stainless steel wiring techniques to lateral mass plates and to polyaxially screw and rod fixation (Figs. 158-6 and 158-7). This has made single-stage anterior-posterior reconstructive procedures more feasible and reasonable in a larger number of patients.¹³

FIGURE 158-6 A case demonstrating the use of a lateral mass plate system, which became the first widely used screw system.

FIGURE 158-7 Patient with severe ossification of the posterior longitudinal ligament treatment with anterior corpectomy with a fibular strut and plate fixation. Posterior polyaxial screw system using lateral mass fixation and C7 pedicle screw fixation.

The surgical training for treatment of complex spinal disorders has also rapidly expanded. As a result, anterior decompression involving segmental reconstructions and corpectomies has become increasingly routine for spinal surgeons.

Technique

Many variations can occur regarding the strategy on planning these cases. Typically, the authors prefer to initiate the approach from the anterior direction first. However, this can be influenced by such things as the need to remove posterior instrumentation. This description will be focused on the typical case of cervical degenerative disease requiring a circumferential decompression.

Imaging

Patients should be evaluated with preoperative plain x-rays in addition to the expected magnetic resonance imaging (MRI) and or computed tomography (CT) myelogram. Abnormal signal on MRI images may also offer some prognostic information regarding potential patient recovery. Mastronardi et al. studied patients with spondylotic myelopathy and found that T2 signal changes are potentially reversible with decompression. T1-weighted signal changes were associated with the worst prognosis and were thought to represent irreversible damage to the cord (Figs. 158-3 and 158-8). Patients with myelopathy that had no signal abnormalities on their preoperative MRI were found to have the best prognosis.¹⁴ This is an interesting concept to include in the debate of whether patients with cervical cord compression should be considered for early treatment or for a conservative approach during follow-up.

FIGURE 158-8 Patient with previous posterior fixation for C3–C4 fracture with sublaminar wires and methymethracrylate presents with juctional stenosis and myelomalacia.

Flexion/extension studies can also be helpful to show the overall degree of sagittal balance, as well as to determine whether a kyphotic deformity is mobile or fixed. Great importance should be given to the cervicothoracic junction if one is considering termination of the construct at C7. It is not infrequently observed that patients may not have significant stenosis at C7–T1, but may have an anterolisthesis. If this is the case, then the construct should extend past the cervicothoracic junction.

Any consideration to instrument the C2 vertebrae should be complemented with a CT including sagittal reconstructions to assess the location of the vertebral artery and its potential risk for injury. If the artery is contained within the pars intra-articularis, then one may want to consider avoiding C2 pars or pedicle screws. A newer option may be to place C2 translaminar screws. This is now much more feasible since the latest generation of polyaxial screw systems have a very wide screw head range of motion.

Anesthesia

Preoperatively the patient should be assessed regarding potential instability, range of motion, and exacerbation of neurologic symptoms associated with neck movement. The findings will determine whether the patient can be put to sleep with standard techniques or may require an awake fiber-optic intubation. This concern cannot be overemphasized in the ankylosing spondylitis patient. This pathology can produce alarming instability with devastating neurologic morbidity. All patients with ankylosing spondylitis should be considered for intubation with awake fiber-optic techniques. The authors believe that it is crucial to discuss the issues with the anesthesia team, so that no miscommunication occurs.

Neuromonitoring

Typically both somatosensory (SSEP) and motor-evoked potentials (MEPs) are used. MEPs monitor the motor pathway of the anterolateral cord and provide rapid feedback to the surgeon about any detrimental changes to these pathways. The latter tend to be very anesthetic sensitive and may require total intravenous anesthesia (TIVA) with propofol. Muscular paralytic agents also cannot be used. Be sure to discuss this with the anesthesiologist prior to the case.

Somatosensory potentials can be used with muscular paralytics and are not quite as sensitive to inhalational anesthetics. The posterior columns are monitored. One potential drawback to SSEPs is that a significant amount of time can elapse before the surgeon is alerted to dysfunction of the tissue.¹⁵

Some patients with severe myelopathy may not have adequate potentials preoperatively. This should be documented in the operative report.

Patient Positioning

Circumferential surgery is highly demanding for surgical, nursing, and anesthesia teams, particularly because of the need to reposition the patient during the procedure. This creates a potential for morbidity to the patient, including IV or endotracheal tube dislodgment, sterile site contamination, spinal instability issues, and even patient falls.

The Jackson table is commonly used to prevent these issues, as it is specifically designed for safely rotating the patient. However, one must be sure that the staff is familiar with the use of the device. A final check of all restraining straps and structural fixation points should be checked by the surgeon before the okay is given to “flip” the patient. Pressure points should be protected with extra care, as these surgeries can be lengthy.

Anterior Approach

Standard operative exposure is used, with surgeon preference dictating a right- or left-sided incision. Much has been passionately debated on the topic of side of surgical approach. Although anatomic studies may be able to demonstrate some theoretical benefits to a left-sided approach, the clinical literature does not seem to reflect a significant difference in outcome.¹⁶ If the patient has had prior surgery, then the same side should be used to avoid a catastrophic bilateral recurrent laryngeal nerve injury. Any need to approach from the opposite side of a previous surgery should be cleared prior to surgery by direct laryngoscopic examination. A vertical incision is usually preferred, as this allows for adequate tissue dissection and exposure of multiple levels. A horizontal incision can certainly be used for less extensive anterior cases. An extensive subplatysmal dissection, along with a wide opening of the superficial cervical fascia will facilitate the exposure.

Once the standard dissection has been completed, self-retaining retractors may be placed. It is a good practice to release the retractors several times to avoid excessive pressure and ischemia to the underlying tissues. The anesthesiologist may also consider deflating and reinflating the endotracheal tube balloon. These maneuvers are believed to reduce patient dysphagia and recurrent laryngeal nerve (RLN) morbidity. Apfelbaum and colleagues reported on a series of 900 consecutive patients over a 12-year period. They began to institute a protocol of deflating the cuff of the endotracheal tube so that it would be recentered in the larynx. After this protocol was adopted, they observed a drop in the incidence of RLN injury from 6.4% to 1.69%.¹⁷ Multiple levels of surgery and female gender have been shown to increase the risk of dysphagia at 6 months follow-up.^18,¹⁹

Once proper exposure has been obtained, then the appropriate decompression may be carried out. Whenever possible the authors prefer to perform segmental reconstruction via multiple diskectomies rather than corpectomies. This helps to ensure adequate creation of lordosis, as well as improving load sharing of the anterior column. When this technique does not allow for adequate decompression, then corpectomies are used. If the extent of decompression requires the use of corpectomy, then attempts should be made to create lordosis. The authors commonly do this by cutting a fibular shaft allograft on an angle. Structural cages are also manufactured with angled end pieces to contribute to lordosis.

Anterior plating is commonly used by most surgeons to encourage fusion and prevent graft subsidence. Anterior instrumentation is not used by the authors in cases without obvious, overt instability, so as to reduce possible dysphagia and avoid load shielding of the anterior column.²⁰

Closure is performed with the utmost attention to hemostasis. The importance of this step cannot be over emphasized as the patient will be in the prone position to complete the rest of the procedure. This will always create some degree of soft tissue edema. Even minor bleeding can then in turn create a very significant airway problem for the patient. We also typically place an anterior drain for the same reason.

Repositioning of the Patient

A Jackson table is commonly used as an efficient method to “flip” the patient. However, certain precautions should be taken by the surgical team when using this device. It is always prudent to make sure that all members of the team are familiar with the table and its system of locks and straps. Significant shifting can occur and should be evaluated as soon as the table is in the locked position. One final point is that the endotracheal tube should always be disconnected by the anesthesiologist prior to the “flip.”

Posterior Approach

A standard midline incision is used with a careful subperiostial takedown of the paraspinous muscles. Attention to meticulous technique at this point in the procedure can help to avoid needless blood loss. Dissection of the lower cervical/upper thoracic area should be done in a manner that prevents working in a hole, as there are usually large veins that can be frustrating to control. Proper positioning that avoids increasing intra-abdominal or intrathoracic pressure will also help reduce distention of these veins.

Decompression is performed via a standard laminectomy. Most cases of cervical spondylitic myelopathy only require removal of the roof of the C7 lamina. The authors typically preserve the rest of the lamina, along with the C7 spinous process. This seems to allow terminating the construct at the C7 pedicle. Care is also taken to preserve the muscular attachments to the C7 spinous process. Muscular preservation has generally received more attention in relation to laminoplasty, but the authors believe that it is also important to avoid junctional breakdown at the cervicothoracic region.²⁰ If there is any doubt of the structural integrity of the C7-T1 level, then one should consider incorporating the thoracic spine into the construct.

Instrumentation of the posterior elements is typically performed with standard lateral mass instrumentation from C3–C6. Screw implantation is generally preferred with a Magerl or similar technique.²¹ For the termination of the construct, where the forces will be the highest, the authors prefer to use the pedicle of C7, as it generally allows for a very strong fixation point.²² We have also observed that the standard 3.5-mm screws are susceptible to fracture at the neck of the screw. This complication has not generally been observed since we started using 4.0-mm diameter screws. Although the C7 pedicle can be found using external landmarks, the use of internal landmarks can reduce the rate of perforation with minimal increase in intraoperative time.²³ A laminotomy is performed on the superior aspect of the C7 lamina, along with a C6–C7 foraminotomy. A micro ball–tipped probe is then used to palpate the C7 pedicle. This region can be associated with a significant epidural venous plexus. Gelfoam and thrombin or Flo-Seal are generally kept on the surgical field to minimize blood loss. There is a significant downward angle of the C7 pedicle relative to the C6 lateral mass. Because of this we commonly skip instrumentation of the C6 lateral mass, in order to properly use the pedicle of C7.

Complications

The actual risk of performing circumferential surgery has been somewhat nebulous as the technique has expanded. Several papers have attempted to detail the morbidity to the patient. Gok et al. retrospectively reviewed lone anterior cases versus combined anterior posterior cases and found a similar rate of complications.¹³ Boakye et al. completed an extensive review of 58,115 admissions of patients with cervical spondylitic myelopathy from 1993 to 2002. While this study did not separate out specifics of circumferential surgery, it does contain some interesting data. The overall mortality rate was 0.6%. The addition of just one postoperative complication increased the mortality rate by 20-fold. This also translated into an average increased stay of 4 days and $10,000 in additional charges. Patients older than 84 years of age had a 40-fold increase in adverse outcomes.²⁴

The anterior cervical approach is well known to be associated with postoperative dysphagia. The actual incidence is somewhat more uncertain. Studies demonstrate a significant range from 12.5% to nearly 50% for short-term complications. Long-term issues also range from roughly 10% to 31.5%.

Explanations for dysphagia following the anterior cervical approach include direct pressure to the esophagus, denervation injury, and local tissue ischemia. Histologic studies also show that the development of fibrosis within the esophageal wall may play an important role. Injury to the recurrent laryngeal nerve has also been associated with dysphagia, but appears to occur at an incidence of 0.07% to 11%. Because the right nerve also innervates the esophagus, it has been suggested that the left approach is preferred.²⁵^–²⁷

Dysphonia can is also associated with the anterior cervical approach. This risk increases with multilevel procedures. The higher the level of the cervical spine also has a transition of anatomy to the more delicate tissues of the retropharyngeal space. Because circumferential cases involve the patient commonly being placed in the prone position for a significant length of time after the completion of the anterior approach, edema and hematomas can be a concern. The authors generally recommend the use of an anterior drain and intraoperative corticosteroids. Airway compromise can also occur because of these issues. The appearance of significant neck and facial edema should prompt a low threshold for the patient to remain intubated. Sagi et al. have shown that there is an increased risk of airway complication with anterior cervical surgery in patients who have undergone multilevel surgery, operative time greater than 5 hours, and blood loss of greater than 300 ml. This obviously describes the typical patient for an anterior posterior reconstruction. This paper also showed that the mean time for these problems to develop was 36 hours postoperatively.²⁸ It is the authors’ observation that the patients undergoing anterior posterior reconstructions tend to develop these complications early in the postoperative period. This is probably because of the time that the airway was in the dependent position during the posterior portion of the procedure. Obviously, one should have a low threshold for concern and action over issues of upper airway difficulty in these patients. Direct observation along with lateral x-rays generally provide the best assessment of the patient.