Complications Associated with Dorsal Cervical Instrumentation

David M. Jackson

Louis F. Amorosa

John G. Heller

Steven C. Ludwig

Subaxial dorsal cervical instrumentation methods have been used for more than a century. Today, several technologies are available for fixation of the cervical vertebrae. Depending on circumstances, a surgeon can use wires, cables, hooks, or screws to gain control of a vertebral segment. These points of purchase can then be connected to one another with rods, plates, or special plate-rod hybrids. A detailed understanding of the morphometry of the dorsal cervical bone elements has allowed surgeons to safely undertake more complex reconstruction techniques in the subaxial cervical lateral masses and pedicles.

Dorsal instrumentation and fusion generally are used in cases of cervical instability secondary to trauma, rheumatoid arthritis, ankylosing spondylitis, neoplastic disease, infections, and degenerative conditions (1). When properly applied, segmental instrumentation techniques can provide sufficient stabilization until fusion occurs; however, as with any surgical technique, they are not without risks of complications. The most formidable complications associated with subaxial cervical fixation include vascular injury, neurologic injury, hardware failure, and infection. Despite these potential pitfalls, when meticulous preoperative planning has defined the pertinent anatomic boundaries, the techniques are straightforward and the complications are few.

Subaxial dorsal cervical fixation began with wiring techniques that ultimately included various ways to grasp the spinous processes, laminae, and facet joints. Omeis et al. (2), Hadra introduced interspinous wiring for the treatment of dorsal ligamentous instability in 1891. Advantages of wire fixation continue to include low cost and technical ease, considering that wire placement requires minimal radiographic guidance (3, 4, 5, 6, 7, 8, 9, 10, 11, 12 and 13). However, wire fixation is unable to resist forces other than pure flexion and is of limited usefulness when the dorsal elements are removed or compromised. External bracing requirements tend to be greater than for current segmental fixation methods. Additionally, wiring methods were not well suited to cases with vertebral body insufficiency because they were not able to withstand ventral compressive forces and resultant kyphosis.

The main risks associated with wiring techniques are excessive intrusion into the spinal canal with secondary neurologic and dural injury and wire cutout from transverse process or lamina with subsequent loss of fixation. Overtightening of the wires can also induce hyperextension, leading to spinal canal or neuroforaminal stenosis. The incidence of all these complications can be reduced with meticulous surgical technique (12,14,15).

Unlike newer titanium screw and rod constructs, stainless steel wires are more likely to interfere with postoperative magnetic resonance imaging (MRI). The introduction of titanium cables markedly improved postoperative imaging artifact. However, the braided titanium cables are somewhat more awkward to work with in the depths of a wound. Lateral mass screws, introduced in 1979 by Roy-Camille et al. (16) and later modified by Anderson et al. (17) and Jeanneret et al. (18), and pedicle screws championed by Abumi et al. (19, 20, 21, 22 and 23) have made wiring techniques uncommon in contemporary dorsal cervical procedures (24, 25, 26, 27 and 28).

Lateral mass screws initially were used with plates and offered the advantages of greater immediate stability and stronger fixation compared with standard wiring techniques (2,16, 17 and 18). Segmental screw fixation made dorsal cervical arthrodesis for challenging pathomechanical situations more likely to succeed and also led to a reduction in the need for postoperative halo vest immobilization. Plates, by virtue of their geometry and hole-to-hole spacing, dictated the starting points for screws, which, in some cases, create technical challenges.

The subsequent evolution of flexible-head screw and rod systems was another substantial improvement. Multiple studies have shown the advantages of screw and rod systems over plate constructs (2,24,29). Rods can be more readily contoured and allow screw placement to be dictated by the dorsal anatomy as opposed to the constraints of the plating system. In addition, rod-screw constructs allow compression, distraction, and lateral rotatory reduction techniques, and can easily be extended across the occipitocervical or cervicothoracic junctions. Screw back out and implant failure are less common with rods, although the incidence of neurovascular complications remains comparable to those encountered with plating constructs (24,26,29).

Abumi et al. (19) initially proposed the placement of cervical pedicle screws as a way of dealing with the poor fixation and attenuated lateral masses of C2 and the thin lateral mass of C7. These authors went on to adapt this technique to instrumentation of the entire subaxial spine. Since then, Abumi et al. (20, 21, 22 and 23) have presented several retrospective studies of the treatment of patients with pedicle screw fixation for traumatic, neoplastic, degenerative, and deformity conditions of the cervical spine, reporting low complication rates.

COMPLICATIONS ASSOCIATED WITH SUBAXIAL DORSAL CERVICAL INSTRUMENTATION

VASCULAR INJURY

One of the most serious complications associated with dorsal cervical instrumentation is injury to the vertebral artery. Vertebral artery injury can be associated with acute and late onset hemorrhage, thrombosis, embolism, fistula, pseudoaneurysm formation, cerebral ischemia, and death. The highest reported incidence (4.1% to 8.2%) of vertebral artery injury during cervical spine surgery is associated with C1-C2 transfacet screw placement (30). However, in the subaxial cervical spine, such injuries seem to be far less common. In fact, the current literature has no reports of vertebral artery injury with subaxial lateral mass screw placement.

In their studies of cervical pedicle screw placement, Abumi et al. (19, 20, 21, 22 and 23) reported an incidence of pedicle breach ranging from 5.3% to 6.7%, although none resulted in clinical complications involving the vertebral artery. Other authors have tried to duplicate the results achieved by Abumi et al. using cervical pedicle screws in the subaxial cervical spine, with variable results. One study (31) reported on 94 subaxial cervical pedicle screws in 26 patients. Only 66 screws (70%) did not breach the pedicle at least partially. Eight screws (9%) critically breached the foramen transversarium, with one patient experiencing temporary paresis and another experiencing sensory loss with incomplete recovery after revision surgery. Another study (32) compared the use of subaxial cervical pedicle screws by using a standard lateral view fluoroscopy guidance technique in one group and a computer navigation system combined with percutaneous instrumentation in another group. In the conventional group, 93 screws were placed in 20 patients and resulted in a 9% pedicle breach rate. The computer-assisted group of 167 screws in 32 patients experienced a 3% breach rate. No neurologic or vascular complications occurred in either group.

A thorough understanding of the anatomic course of the vertebral artery and its variations is essential to preserving its integrity during dorsal instrumentation. An anomalous course of the vertebral artery can further increase a patient’s risk of injury. This is most often an issue in the atlantoaxial region, where up to 20% of the population can have anomalous positioning of the vertebral artery (30). In the subaxial spine, a tortuous vertebral artery can erode into the pedicles and vertebral bodies, putting the artery at risk if pedicle screw placement is attempted (30). In a cadaveric study, Curylo et al. (33) identified an ectopic vertebral artery in (2.7%) of 222 specimens. The anomalies were identified by comparing the medial aspect of the uncovertebral joint with the medial aspect of the transverse foramen, with normal being defined as a foramen greater than 1.5 mm lateral to the uncovertebral joint. All six anomalies were unilateral, with five being on the left and one being on the right. The relationship of the vertebral artery also varies at C7. In the majority of individuals, the first segment of the vertebral artery arises from the subclavian artery and courses ventral to C7, entering the foramen transversarium of C6. However, in a small subset of patients, the vertebral artery enters directly into the C7 foramen transversarium. Careful scrutiny of the preoperative axial view computed tomographic (CT) scan at the C7 level for a dysplastic foramen transversarium typically allows for safe placement of a pedicle screw at the C7 level. On the other hand, if the C7 foramen transversarium appears large and well formed, the surgeon should assume that the vertebral artery has entered the cervical spine at that level. One cadaveric study (34) found that 3 of 40 cadavers (8%) had a vertebral artery that at least partially entered the foramen transversarium of C7. Another study (35) put this incidence at 5%. If the surgeon is concerned about potential risk to a vertebral artery during dorsal instrumentation, a preoperative CT angiogram or magnetic resonance angiogram can be obtained to more precisely determine its path.

Only gold members can continue reading. Log In or Register to continue