23 Navigated and Robotic Anterior Odontoid Peg Fracture Fixation

10.1055/b-0039-172734

23 Navigated and Robotic Anterior Odontoid Peg Fracture Fixation

Kartik Shenoy and Ali Bydon

Abstract:

Odontoid peg, or dens, fractures are being recognized more frequently due to the increasing use of computed tomography in the traumatically injured patient and consequently, there is an increased rate of operative intervention. Transverse or oblique fracture patterns are amenable to anterior fixation; however, precise placement of anterior odontoid screws is critical to achieving fracture union. Traditionally, biplanar fluoroscopy was used to place anterior odontoid screws, but the setup can be cumbersome and there is increased radiation exposure to the patient and surgeon. Navigation and robotics have allowed for more accurate placement of anterior odontoid screws while minimizing the amount of radiation, however the long-term clinical outcomes are still to be determined.

23.1 Introduction

Odontoid fractures can be the result of spinal trauma and if missed can be life-threatening. Fractures of the odontoid peg, or dens, constitute 5 to 15% of all cervical spine injuries. ¹ In 1928, Osgood and Lund reviewed 55 cases and found the mortality rate to be more than 50%. ² Advances since then as described by Amyes and Anderson in a 1956 report of 63 cases had a mortality rate of 8% with a 5% rate of nonunion. ¹ The high rate of mortality is largely a result of missed injuries, as the upper cervical spine is difficult to visualize on plain radiographs given the overlapping osseous structures. They are frequently missed due to the lack of clinical symptoms other than neck pain. With the advent of computed tomography (CT) and its application in the trauma setting, fractures are better identified and characterized. ³

According to the Anderson-D’Alonzo classification, there are three types of odontoid peg fractures based on the fracture pattern from cephalad to caudal. Type I is a fracture of the tip usually due to an avulsion of the alar ligament. Type II is a more caudal fracture through the waist or base where the odontoid peg meets the body of the axis. The fracture line in a type III extends into the body and has a larger fracture bed surface area. ⁴ Type II fractures are typically treated operatively with either C1–C2 posterior fusion or anterior odontoid screw fixation. C1–C2 posterior fusion is performed most commonly with C1 lateral mass with C2 pedicle screws. ⁵ ^, ⁶ ^, ⁷ However, given that the C1–C2 articulation contributes greatly to cervical rotation, postfusion, the patient may lose up to 50% of neck rotation.

Anterior fixation with one or two odontoid screws is another treatment option; unlike a fusion, it offers the advantage of preserving rotation. In order to perform anterior fixation, the fracture pattern must be transverse or in the anterosuperior to posteroinferior orientation and reducible without significant comminution. ⁸ The surgeon must also be mindful of the patient’s chest size, as insertion of the screw at the correct angle may be limited by a larger, barrel-shaped chest. When anterior fixation is appropriate, imaging using fluoroscopy or intraoperative CT is suggested for proper screw placement.

23.2 Traditional Surgical Technique

Anterior odontoid osteosynthesis can be performed using two biplanar fluoroscopy machines to obtain images of the odontoid peg in the true anteroposterior (AP) and lateral planes. Prior to incision, using halo traction, a Mayfield or manual reduction, reduction of the fracture should be attempted. With the patient in the supine position, the head is extended to facilitate reduction and screw insertion. Before making incision, a K-wire is placed along the neck in the intended direction and then confirmed on fluoroscopy to guide placement of a transverse, anteromedial incision. An anterior approach to C2–C3 is made via an incision at approximately the C5–C6 level. The platysma is divided and blunt dissection medial to the sternocleidomastoid is performed down to the prevertebral fascia. The fascia is then incised at the C2–C3 level and the disk space is identified. A partial removal of the anterosuperior edge of C3 body is performed to allow exposure of the inferior edge of C2 and the screw start point. A K-wire is then advanced under fluoroscopy and then a 4.5-mm cannulated cancellous screw is placed over the wire. ⁹

23.3 Navigation

Navigation for anterior odontoid screw placement evolved through the need for more accurate screw placement and to minimize radiation exposure and complications. Although screws can be placed via fluoroscopy as described earlier, there are still complications such as screw cutout, vascular injury, or even neurological injury at a rate of 0.2 to 5%. ¹⁰ ^, ¹¹ The need for accurate placement of screws led to the field of spinal navigation, and technology that was once used for precision pedicle screw placement is now used in anterior odontoid screw placement. There are a variety of navigation technologies available and they vary based on the imaging modality: fluoroscopy-based, three-dimensional fluoroscopy, and intraoperative CT.

23.3.1 Fluoroscopy-Based Computer Navigation

Fluoroscopy-based computer navigation, or virtual fluoroscopy, as described by Battaglia et al is a technique that employs the traditional two-dimensional C-arm images in conjunction with an optical tracking system and computer to allow real-time visual tracking of screw trajectory relative to preacquired images. ¹² In this technique, standard AP and lateral fluoroscopic images are obtained and then instruments are calibrated and synchronized to a computer and a reference frame attached typically to a Mayfield frame. The software is then able to show predicted positions of instruments on the fluoroscopy screen in order to provide real-time feedback to the surgeon as instruments are positioned. According to Chibbaro et al, virtual fluoroscopy allows for surgeons to perform a high-risk surgery in a safer, easier, and faster way. ¹³ Although there is a learning curve, there is a significant reduction in fluoroscopy time which reduces radiation exposure to both the patient and the surgeon as a total of only four images are needed: AP and lateral pre- and postoperatively. These findings were also supported by Battaglia et al in which they directly compared traditional fluoroscopy to virtual fluoroscopy. ¹²

23.3.2 Three-Dimensional Image-Based Navigation and Computed Tomography

The isocentric C-arm three-dimensional fluoroscopy is a form of intraoperative fluoroscopy and CT to provide standard fluoroscopic images as well as cross-sectional imaging in three planes allowing increased accuracy during surgery. A set of defined consecutive images are acquired in a 190-degree orbital plane and then multiplanar image reconstruction is performed to obtain real-time axial, sagittal, and coronal imaging of the cervical spine. Similar to the fluoroscopy technique detailed earlier, a reference frame is placed on a Mayfield or similar head clamp. After calibration, the trajectory of screws or instruments can be viewed in real time on the three-dimensional cross-sectional imaging. Summers et al reported a case series of nine patients who underwent anterior odontoid screw placement using isocentric C-arm three-dimensional fluoroscopy and found that all screws were accurately placed, and no additional operative time was needed. ¹⁴ Furthermore, using the three-dimensional fluoroscopy, they were able to perform an immediate postoperative CT scan to ensure proper screw placement prior to leaving the operating room.

Many ensuing studies have investigated the role of three-dimensional fluoroscopy and found similar results. ¹⁵ ^, ¹⁶ Additional studies have directly compared three-dimensional fluoroscopy to the traditional technique and found a significantly less fluoroscopy time when using three-dimensional fluoroscopy without any difference in operating time, blood loss, or complications. ¹⁷ Martirosyan et al reported that total time in operating room was not statistically different in either group as the three-dimensional fluoroscopy group had shorter operative time but longer setup time, whereas the traditional fluoroscopy group had longer operative time but shorter preoperative setup. ¹⁸ In this study, they also reported on outcomes and found that patients in the three-dimensional fluoroscopy group had higher outcome scores and fusion rates without any difference in the rate of complications.

Intraoperative CT using an O-arm (Medtronic, Minneapolis, MN) can also be used with navigation-based placement of anterior odontoid screws. ¹⁹ In 2017, Pisapia et al presented the first report on using an O-arm for anterior odontoid screws in eight patients. ²⁰ This technique also employs a reference point to be placed on the head clamp. After performing dissection, the O-arm is brought into the operating field and a full spin is performed to obtain cross-sectional imaging. A calibrated probe can be used to plan the start point and trajectory of the screw (Fig. 23‑1 and Fig. 23‑2). Next, using a drill which is also registered to the system, the planned pathway is drilled (Fig. 23‑3). Real-time fluoroscopy can then be performed while drilling to ensure that the drill matches the planned path (Fig. 23‑4). The ability of the O-arm to capture AP and lateral fluoroscopic images while advancing the drill and screw is not a necessity but can be used to double check positioning. This is an advantage, as there is no movement of the O-arm when transitioning from an AP to lateral view as would be performed with traditional or three-dimensional fluoroscopy which could potentially get in the surgeon’s way. Final intraoperative O-arm can confirm accurate screw placement (Fig. 23‑5). More studies utilizing the O-arm for anterior odontoid screw placement are needed and further studies comparing virtual fluoroscopy to three-dimensional fluoroscopy and O-arm with regard to outcomes and complications would allow for the safest, easiest, and most cost-effective technique to be identified.