Principles Of Image-Guided Spinal Navigation

The use of an image-guided navigational system for localizing intracranial lesions has been previously described.^7,8 Image-guided navigation establishes a spatial relationship between a preoperative CT image and its corresponding intraoperative anatomy. Both the CT image and the anatomy can each be viewed as a three-dimensional coordinate system with each point in that system having a specific x, y, and z Cartesian coordinate. Using defined mathematical algorithms, a specific point in the image data set can be matched to its corresponding point in the surgical field. This process is called registration and represents the critical step of image-guided navigation. At least three points need to be matched, or registered, to allow for accurate navigation.

A variety of navigational systems have evolved during the past decade. The common components of most of these systems include an image-processing computer workstation interfaced with a two-camera optical localizer (Fig. 305-1). When positioned during surgery, the optical localizer emits infrared light toward the operative field. A handheld navigational probe mounted with a fixed array of passive reflective spheres serves as the link between the surgeon and the computer workstation (Fig. 305-2). Alternatively, passive reflectors may be attached to standard surgical instruments. The spacing and positioning of the passive reflectors on each navigational probe or customized trackable surgical instrument is known by the computer work-station. The infrared light that is transmitted toward the operative field is reflected back to the optical localizer by the passive reflectors. This information is then relayed to the computer work-station, which can then calculate the precise location of the instrument tip in the surgical field as well as the location of the anatomic point on which the instrument tip is resting.

FIGURE 305-1 Image-guided navigational workstation with infrared camera localizer system.

FIGURE 305-2 Navigation probe and drill guide for spinal surgery.

The initial application of navigational principles to spinal surgery was not intuitive. Early navigational technology applied to intracranial surgery used an external frame mounted to the patient’s head to provide a point of reference to link preoperative image data to intracranial anatomy. This was not practical for spinal surgery. The current generation of intracranial navigational technology uses reference markers or fiducials that are attached to the patient’s scalp before imaging. However, the use of these surface-mounted fiducials for spinal navigation is not practical because of accuracy issues related to a greater degree of skin movement over the spinal column.^9,10 This is less of a problem with intracranial applications because of the relatively fixed position of the overlying scalp to the underlying anatomy.

The application of navigational technology to spinal surgery involves using the rigid spinal anatomy as a frame of reference. Bone landmarks on the exposed surface of the spinal column provide the points of reference necessary for image-guided navigation. Specifically, any anatomic landmark that can be identified intraoperatively as well as in the preoperative image data set can be used as a reference point. The tip of a spinous or transverse process, a facet joint, or a prominent osteophyte can serve as a potential reference point (Fig. 305-3). Because each vertebra is a fixed, rigid body, the spatial relationship of the selected registration points to the vertebral anatomy at a single spinal level and is not affected by changes in body position.

FIGURE 305-3 Navigational workstation screen demonstrating a paired point registration plan for the insertion of T12 pedicle screws. Three discreet bony landmarks are selected at the T12 level. In this case, the lateral margins of the two T12 transverse processes and the tip of the T12 spinous process have been selected.

The purpose of the registration process is to establish a precise spatial relationship of the image data with the physical space of the patient’s corresponding surgical anatomy. If the patient is moved after registration, this spatial relationship is distorted, making the navigational information inaccurate. This problem can be minimized by the optional use of a spinal tracking device consisting of a separate set of four passive reflectors mounted in a known configuration on a small frame. This reference frame can be attached to the exposed spinal anatomy and its position in space tracked by the infrared camera system (Fig. 305-4). Movement of the spinal anatomy and the attached frame alerts the navigational system, which can then make the appropriate correctional calculations to maintain accuracy and eliminate the need to repeat the registration process. The disadvantage of using a tracking device is the added time needed for its attachment to the spine, the need to maintain a line of sight between it and the camera, and the inconvenience of having to perform the procedure with the device placed in the surgical field. It is particularly cumbersome when image-guided navigation is used during cervical procedures. Alternatively, image-guided spinal navigation can be performed without a tracking device.^1,12 This involves acknowledging the effect of patient movement on the accuracy of image-guided navigation and maintaining reasonably stable patient position during the relatively short amount of time needed (i.e., 10 to 20 seconds) for the selection of each appropriate screw trajectory. Patient movement can potentially occur with respiration, from the surgical team leaning on the table, or from a change of table position. Movement associated with patient respiration is negligible and does not require any tracking, even in the thoracic spine. Although movement associated with leaning on the table or repositioning the table or the patient will affect registration accuracy, it can be easily avoided during the short navigational procedure. If inadvertent patient movement does occur, the registration process can be repeated.

FIGURE 305-4 Reference frame attached to a spinous process in the surgical field. The reference frame monitors inadvertent movement of the spinal anatomy that may affect navigational accuracy.

Three different registration techniques can be used for spinal navigation: paired point registration, surface matching, and automated registration. Paired point registration involves selecting a series of corresponding points in a computed tomography (CT) or magnetic resonance imaging (MRI) data set and in the exposed spinal anatomy. The registration technique is performed immediately after surgical exposure and before any planned decompressive procedure. This allows for the use of the spinous processes as registration points.

A specific registration point in the CT image data set is selected by highlighting it with the computer cursor. The tip of the probe is then placed on the corresponding point in the surgical field, and the reflective spheres on the probe handle are aimed toward the camera. Infrared light from the camera is reflected back, allowing the spatial position of the probe’s tip to be identified. This initial step of the registration process effectively links the point selected in the image data with the point selected in the surgical field. When at least three such points are registered, the probe can be placed on any other point in the surgical field, and the corresponding point in the image data set will be identified on the computer workstation.

Alternatively, a second registration technique called surface matching can be used. This technique involves selecting multiple, random (nondiscrete) points on the exposed surface of the spine in the surgical field. This technique does not require prior selection of points in the image set, although several discrete points in both the image data set and the surgical field are frequently required to improve the accuracy of surface mapping. The positional information of these points is transferred to the work-station, and a topographic map of the selected anatomy is created and matched to the patient’s image set.¹¹

Typically, paired point registration can be done more quickly than surface mapping. The average time needed for paired point registration is 10 to 15 seconds. The time needed for surface mapping is much longer, with difficult cases requiring as much as 10 to 15 minutes. With the need to perform several registration processes during each surgery, this time difference can significantly affect the length of the navigational procedure and the surgery.¹²

More recently, the development of intraoperative CT systems that incorporate navigational technology has allowed for the use of an automated registration technique. This approach involves attachment of a reference frame on the exposed spinal anatomy. The intraoperative imaging system is then positioned, and a CT image of the pertinent spinal anatomy obtained. A second reference frame attached to the CT scanner allows its position relative to the spine reference frame and, therefore, the spinal anatomy to be tracked as well. With these spatial relationships established, the navigational system can then perform an automated registration of the spinal anatomy eliminating the manual efforts required of the surgeon with the paired point and surface matching techniques. The scanner is removed, and real-time navigation of up to five separate spinal levels can be performed.

After accurate registration, a trackable probe with attached reflective spheres can be positioned on any surface point in the surgical field. As the probe is tracked by the camera system, three separate reformatted CT images centered on the corresponding point in the image data set are immediately presented on the workstation monitor. Each reformatted image is referenced to the long axis of the probe. If the probe is placed on the spinal anatomy directly perpendicular to its long axis, the three images will be in the sagittal, coronal, and axial planes. A trajectory line representing the orientation of the long axis of the probe will overlay the sagittal and axial planes. A cursor representing a cross section through the selected trajectory will overlay the coronal plane. The insertional depth of the trajectory can be adjusted to correspond to selected screw lengths. As the depth is adjusted, the specific coronal plane will also adjust accordingly, with the position of the cursor demonstrating the final position of the tip of a screw placed at that depth along the selected trajectory. As the probe is moved to another point in the surgical field, the reformatted images, as well as the position of the cursor and trajectory line, will also change. The planar orientation of the three reformatted images will also change as the probe’s angle relative to the spinal axis changes. When the probe’s orientation is not perpendicular to the long axis of the spine, the images are displayed in an oblique, or orthogonal, plane. Regardless of the probe’s orientation, the navigational workstation will provide the surgeon with a greater degree of anatomic information than can be provided by any intraoperative imaging technique.

Clinical Applications

Image-guided spinal navigation was initially evaluated for the insertion of pedicle screws in the thoracic and lumbosacral spines of cadaver specimens. The accuracy of screw insertion was documented by plain film radiography and thin-section CT imaging of the instrumented levels. Satisfactory screw placement was noted for 149 of 150 inserted screws.² The initial clinical application of image-guided spinal navigation was its use for lumbosacral pedicle fixation.^1,13,14 Other spinal applications gradually evolved including transoral decompression, cervical screw fixation, thoracic pedicle fixation, decompression of spinal metastasis, and anterior thoracolumbar decompression and fixation procedures.^12,15–20

The application of image-guided navigation to spinal surgery is directed by the complexity of the procedure and, specifically, by the need to visualize the unexposed spinal anatomy. Image-guided navigation can be used with or without standard intraoperative imaging techniques (i.e., fluoroscopy). In either case, image-guided navigation provides the surgeon with an improved orientation to the pertinent spinal anatomy, which subsequently facilitates the accuracy and effectiveness of the procedure.

Pedicle Fixation

Pedicle fixation has gained acceptance as an effective and reliable method of spinal stabilization. However, because of the variations of pedicle anatomy within each patient, the safe and precise placement of pedicle screws can be difficult. Suboptimal screw placement can result in varying degrees of neural injury and fixation failure. These complications can be minimized if the surgeon is provided with accurate spatial orientation to each pedicle to be instrumented before screw insertion.

Image-guided spinal navigation can be used in place of fluoroscopy to assist in the insertion of pedicle screws in both the thoracic and lumbosacral spine. Although fluoroscopy provides real-time imaging of spinal anatomy, the views generated represent only two-dimensional images of a complex three-dimensional structure. Manipulation of the fluoroscopic unit can reduce this problem, but these maneuvers can be cumbersome and time-consuming. Other disadvantages include the radiation exposure and the need to wear lead aprons during the procedure. Fluoroscopy cannot provide a view of the spinal anatomy in the axial plane. It is this axial view provided by image-guided navigation that makes it superior to fluoroscopy for spinal screw fixation procedures.

The application of image-guided navigation to the spine involves obtaining a preoperative CT scan through the appropriate spinal segments to be instrumented. The images consist of a three-dimensional volume data set of contiguous axial CT images. Alternatively, MRI data may also be used. The image is then transferred to the computer workstation through an optical disk or a high-speed data link. If paired point registration is to be used, three to five reference points for each spinal segment to be instrumented are selected and stored in the image data set.

Intraoperatively, a standard exposure of the spinal levels to be instrumented is performed. A lateral radiograph can be obtained to confirm the appropriate level. The computer workstation and camera localizer are then positioned. The infrared camera detector is mounted at the foot of the table and aimed rostrally for thoracic and lumbosacral procedures.

Image-guided navigation is typically used before any planned decompression in order to use the intact posterior elements as registration points. The first spinal segment to be instrumented is registered using either the paired point or surface mapping technique. When the registration process has been completed, the navigational workstation will calculate and display a registration error (expressed in millimeters) that is directly dependent on the surgeon’s registration technique. The error presented does not represent a linear error but rather a volumetric calculation comparing the spacing of registration points in the surgical field to the spacing of the corresponding points in the image data set. This figure is, at best, a relative indicator of accuracy.

A better method of ensuring registration accuracy is the verification step. This step is typically performed immediately after completing either registration process. The surgeon places the navigational probe on a discrete landmark in the surgical field. With the navigational system now tracking the movement and position of the probe, the trajectory line and cursor on the work-station screen will move to the corresponding point in the image data set provided that registration accuracy has been achieved. If registration accuracy has not been achieved, the cursor and trajectory line may rest on a point other than that selected in the surgical field. If this occurs to a significant degree, the registration process needs to be repeated. This step is more of an absolute indicator of registration accuracy and is important to perform before proceeding with navigation.

When an accurate registration of the first spinal level to be instrumented has been verified, standard bony landmarks for pedicle localization are used to approximate the screw entry point. A drill guide is placed on this entry point, and the navigation probe is passed through the guide. The navigational system is activated, permitting tracking of the probe in the surgical field. Three separate reformatted views are displayed on the work-station screen. Each view represents a separate plane passing through the selected point in the surgical field. For most pedicle fixation cases, these views typically consist of a sagittal, an axial, and a coronal reconstruction. A trajectory line referenced to the long axis of the probe is superimposed on the sagittal and axial views. A round cursor, representing a cross section through the selected trajectory, is superimposed on the coronal view. As the probe is moved through the surgical field, the position of the trajectory line and cursor will change accordingly. Both the width of the trajectory line and the diameter of the cursor can be adjusted to match the relative diameter of the pedicle screws to be used. The length of the trajectory line can also be adjusted (Fig. 305-5).

FIGURE 305-5 Workstation screen demonstrating navigation for an L3 pedicle screw.

As the probe is placed on each pedicle entry point, the images on the workstation screen are presented in real time. As the angle of the probe is adjusted in the axial and sagittal planes, the images immediately update to show the corresponding trajectories. The depth of the coronal view can be adjusted to show the cross-sectional anatomy at any point along the selected trajectory. The orientation of each pedicle to be instrumented can be assessed rapidly and accurately. Any errors in trajectory or entry point selection can be determined and corrected by adjusting the position of the probe and the drill guide through which it passes.

When a satisfactory screw entry point and trajectory have been selected, the probe is removed from the drill guide, and a drill (3 mm diameter) is inserted through the guide. A pilot hole is drilled along the selected trajectory. The purpose of using a drill guide is to preserve the physical trajectory and entry point information acquired through the navigation process. Without a drill guide, it may be difficult to precisely position a drill or pedicle probe on the same point and with the same trajectory selected during navigation. When the pilot hole is placed, a sound can be passed down the hole to ensure adequate positioning. Navigation is then performed for the contralateral pedicle and its pilot hole drilled. The process of navigating each spinal level, including registration, accuracy verification, navigation, and pilot hole placement, typically takes no more than 2 to 3 minutes.

For each additional vertebra to be instrumented, a new set of registration points at that level is selected. This method, termed segmental registration, eliminates any potential discrepancy in anatomic orientation that may be related to a change in patient position between the preoperative CT scan and surgery. Because each vertebra is a fixed rigid body, the spatial relationship of the selected registration points to the vertebral anatomy at a single spinal level is not affected by changes in body position.

After all pilot holes have been drilled, they are tapped and the appropriate size screws inserted. C-arm fluoroscopy or serial radiography is not required. Typically, the combined time for both navigation and screw insertion for a two-level lumbar fixation procedure is about 8 to 10 minutes when using a paired point registration technique. This figure can be considerably higher when using a surface mapping technique owing to the greater time it takes to achieve adequate registration with surface mapping.

In addition to screw placement in the large pedicles of the lumbosacral spine, image-guided navigation can also facilitate screw placement into the smaller pedicles of the thoracic spine (Fig. 305-6). The added precision for screw placement into thoracic pedicles greatly expands the fixation options for managing the unstable thoracic spine and cervicothoracic junction.

FIGURE 305-6 Workstation screen demonstrating navigation for a T8 pedicle screw in a patient with mycotic aneurysm of the aorta.

Image-guided navigation can also be used in place of fluoroscopy for placement of interbody cages in the lumbosacral spine. During removal of the intervertebral disk, the navigational probe can be inserted into the evacuated disk space. With the trajectory length set at zero, the three reformatted images displayed provide optimal spatial orientation to the disk space, allowing for precise placement of the cages (Fig. 305-7).

FIGURE 305-7 A, Workstation screen before L5-S1 disk excision for a posterior interbody fusion (probe tip location and trajectory highlighted by arrows). B, Workstation screen after L5-S1 disk excision. The depth within the disk space and the extent of disk removal can be determined before cage or bone graft insertion (probe tip location and trajectory highlighted by arrows).

C1-2 Transarticular Screw Fixation

Instability of the atlantoaxial complex is frequently managed by the placement of fixation screws through the pars interarticularis of C2, across the facet joint and into the lateral mass of C1. The procedure is typically performed bilaterally using fluoroscopic guidance. The potential risks of this procedure include injury to the vertebral artery if the screw is placed too laterally or ventrally, injury to the spinal cord if the screw is placed too medially, and failure to engage the lateral mass of C1 if the screw trajectory is too ventral. The insertion of a screw on either side may be contraindicated if the pars interarticularis of C2 is too narrow.

Although fluoroscopy provides real-time imaging of the relevant spinal anatomy, the two-dimensional images generated may not be sufficient to provide accurate screw trajectory information. Image-guided navigation adds an additional layer of accuracy by generating multiple planes of imaging through the C1-2 anatomy. As with other image-guided navigation procedures, a preoperative CT scan of the relevant anatomy is obtained. The image is transferred to the computer workstation and can be used to create a preoperative screw trajectory plan. A proposed entry point and target can be selected at the C2 and C1 levels, respectively. The image data set can then be manipulated in multiple planes between these two points to demonstrate the position of a screw placed along the selected trajectory. In addition to a sagittal image that demonstrates the same information provided by lateral fluoroscopy, two other images are presented. One of the images lies perpendicular to the sagittal image along the selected trajectory. This image represents an orthogonal view that lies about midway between the coronal and axial planes through the spine. It demonstrates a second view of the selected trajectory.

An additional view demonstrates an image oriented perpendicular to the long axis of the probe and, therefore, the selected trajectory. A cursor superimposed on this image can show the position of the screw tip along the selected trajectory at millimeter increments. By scrolling through this image, the proposed position of the screw along the selected trajectory can be assessed along its entire path. Although this planning technique does not ensure safe screw placement intraoperatively, it can alert the surgeon preoperatively to avoid screw placement in patients with insufficient anatomy and to select an alternative approach.

Intraoperatively, the patient is positioned, and the posterior C1-2 complex is exposed. A cable and bone graft stabilization procedure at the C1-2 level is performed before navigation and screw insertion. Performing this step first minimizes any independent motion between C1 and C2 during navigation and makes tap and screw insertion easier. If a reference frame is used, it is typically attached to the spinous process of C2.

After placement of the graft and cable, three to five registration points are selected at the C2 level. It is not necessary to include registration points at C1. Although the spatial relationship of C1 and C2 may change between the preoperative scanned position and the intraoperative position, the ability of image-guided navigation to facilitate accurate screw placement is not significantly affected. The technical difficulty of this procedure is the accurate passage of the screw through the narrow pars interarticularis of C2. The lateral mass of C1 is a relatively large target that can be easily reached, provided that there is a reasonably acceptable realignment of C1 and C2 as well as an optimal positioning of the screw within the appropriate C2 anatomy. Although the relative position of C1 and C2 in both the preoperative image set and the surgical field is important, it is not critical enough to interfere with the process of image-guided navigation.

Two separate stab incisions are made on either side of the midline at the C7-T1 level. A drill guide is placed through one of the stab incisions and passed through the paravertebral musculature and into the operative field. A small divot is drilled at the proposed entry site to provide for secure placement of the drill guide. The registration process is performed at the C2 level and its accuracy confirmed using the verification step. The probe is passed through the drill guide, and as its position is adjusted in the surgical field, the images on the workstation screen will adjust accordingly to show the corresponding trajectory in two separate planes and the projected location of the screw tip in the third plane. Orientation to the correct screw position can be assessed rapidly and accurately (Fig. 305-8). Any errors in trajectory or entry point selection can be determined and corrected by adjusting the position of the probe and the drill guide through which it passes. When the correct screw insertion parameters have been selected, the probe is removed from the drill guide, and a drill is inserted. A hole is drilled along the selected trajectory, tapped, and the appropriate length screw inserted. The process is repeated on the opposite side.

FIGURE 305-8 Workstation screen demonstrating a trajectory for insertion of a C1-2 transarticular screw. The lower right screen shows the trajectory in the sagittal plane. The lower left screen represents an orthogonal plane lying between the axial and coronal planes. It conveys the medial-lateral trajectory. The upper left screen represents a plane that is perpendicular to the two other images. It demonstrates the location of the screw tip inserted along the selected trajectory at the indicated depth (screw trajectory and tip location highlighted by arrows).

Although image-guided navigation does not guarantee accurate screw placement, it does provide the surgeon with a greater degree of anatomic information than fluoroscopy alone. The addition of fluoroscopy to this navigational technique provides the greatest degree of precision to the procedure. In this case, navigational technology significantly reduces the time of intraoperative fluoroscopic use because it is typically used only to help position the patient preoperatively and as a final check of the selected trajectory in the sagittal plane immediately after the navigational step.

Segmental C1-2 Screw Fixation

As an alternative to transarticular screw fixation, segmental fixation of C1-2 can be used for managing atlantoaxial instability.²¹ The procedure involves placing a screw into each of the two lateral masses of C1 and two screws down the pedicles of C2. The polyaxial screw heads on each side are then connected with rods. Although this approach potentially reduces the risk for injury to the vertebral artery during screw insertion, it does not eliminate the risk. As with the transarticular technique, precise anatomic orientation is required to avoid arterial or neural injury. Image guidance can supplement intraoperative fluoroscopy to provide the necessary orientation for accurate screw insertion.

As with the transarticular screw fixation technique, a preoperative CT scan is obtained. The posterior C1-2 spine is exposed, and a wire and cable fixation procedure is carried out. Registration is first performed at C1 for placement of the C1 lateral mass screws. The three registration points typically used at C1 include the midline posterior tubercle and the bilateral points marked by the junction of the small pedicle of C1 with its lateral mass (immediately above the two exiting C2 nerve roots). Once registered, the correct trajectory into the lateral mass can be displayed on the workstation screen and the screws inserted (Fig. 305-9). To use image guidance for inserting C2 pedicle screws, the same registration points are used at C2 as those used for transarticular fixation (the C2 spinous process and the two lateral margins of the C2-3 facet). The entry point for the screw is more lateral and the trajectory more medially oriented than for a transarticular screw. The navigation probe is placed through a drill guide onto this entry point and the selected trajectory is displayed on the workstation screen. When the correct entry point and trajectory have been selected, the probe is removed, a drill is inserted, and the pilot hole is drilled. The process is then repeated for the other side. The heads of the screws are then connected with two short rods.

FIGURE 305-9 A, Workstation screen demonstrating navigational information for placement of a screw into the lateral mass of C1. B, Workstation screen demonstrating navigational information for placement of a screw into the pedicle of C2.

Transoral Surgery

Transoral decompression of the upper cervical spine typically requires intraoperative fluoroscopy to help maintain proper anatomic orientation during the procedure. Although orientation in the sagittal plane is easy to obtain with fluoroscopy, depth and medial-lateral orientation are more difficult to assess. Image-guided technology can be used to orient the surgeon in multiple planes during transoral surgery.^12,22

Unlike other spinal applications of image guidance, discrete registration points are not readily available during transoral surgery. For this procedure, surface-mounted markers (fiducials) are applied to the patient before obtaining the preoperative CT. Typically, two fiducials are applied to the mastoid processes, and two are applied to the lateral orbital margins or to both malar eminences. The nasal septum and the anterior tubercle of C1 can also be used as registration points.

The patient is positioned in a three-point head holder. The registration process is performed before draping the patient using the surface-mounted fiducials. Because the registration points will not be accessible during the procedure, a reference frame fixed to the three-point head holder is used for transoral navigation. This allows for changes in patient positioning during surgery without the need to re-register.

During the procedure, the probe can be placed into the site of the decompression. Reformatted sagittal, axial, and coronal CT images are immediately generated providing the surgeon with a precise orientation to the pertinent surgical anatomy. In particular, orientation in the axial plane minimizes the risk for lateral deviation toward the vertebral artery during the decompression (Fig. 305-10). If a posterior fixation is indicated following transoral decompression, the same CT image data set can be used for C1-2 screw placement

FIGURE 305-10 Workstation screen demonstrating navigational information during transoral decompression (probe tip location and trajectory highlighted by arrows).

Anterior Thoracolumbar Surgery

Image-guided spinal navigation can be applied to anterior thoracolumbar surgery to help orient the surgeon to the extent of anterior decompression and to facilitate the precise placement of fixation screws. Although the selection of reference points for anterior spinal surgery is limited by the relative lack of prominent bony landmarks on the anterior aspect of the spinal column, the degree of accuracy required is less than that needed for most posterior screw fixation procedures. This degree of accuracy, termed clinically relevant accuracy, will change according to the procedure being performed. It represents the degree of accuracy needed to achieve a particular surgical task. For example, insertion of a C1-2 transarticular screw has a higher clinically relevant accuracy demand than placing an anterior fixation screw across a large thoracic or lumbar vertebral body. In both cases, image-guided navigation provides clinically relevant accuracy more consistently than fluoroscopy alone.

Potential registration points for the use of image-guided navigation in anterior thoracolumbar surgery include selected landmarks on the vertebral end plates, pedicles, head of the rib, and prominent ventral osteophytes. In general, higher registration errors can be tolerated because of the lower accuracy requirements for most anterior thoracolumbar procedures compared with posterior screw fixation procedures. The accuracy verification step performed immediately after registration can further confirm the achievement of clinically relevant accuracy before proceeding with navigation.

During anterior decompression, the probe can be placed into the partially decompressed site to orient the surgeon to the contralateral margin of the spinal column and, more importantly, to the location of the epidural space (Fig. 305-11A). Orientation to tumor margins can also be obtained by placing the probe into the partially decompressed tumor bed. After decompression, image guidance can be used to guide anterior fixation screws across the vertebra at either end of the corpectomy site (Fig. 305-11B).

FIGURE 305-11 A, Workstation screen demonstrating navigation during removal of an L2 metastasis. Orientation to the contralateral side as well as the epidural space can be obtained. B, Workstation screen demonstrating a selected trajectory for an anterior lumbar fixation screw.

Pitfalls Of Image-Guided Spinal Navigation

Although image-guided spinal navigation has proved to be a versatile and effective tool for facilitating complex surgical procedures, it can be prone to several potential problems before and during its use. In general, these pitfalls and errors are related to issues of accuracy, technique, and overall ease of use of the technology during surgery. A thorough understanding of these potential problems is required to ensure the efficient and effective use of image-guided navigation for spinal surgery.

Like any other computer-based technology, image-guided navigation is highly dependent on the quality of the information imported into the system. Although obtaining the properly formatted CT images and having them correctly transferred to the navigational workstation is important, the critical step of image guidance is the registration process. If the surgeon takes too casual an approach to registration, inaccurate information will be displayed during intraoperative navigation.

Another important principle of image guidance is the understanding that the navigational information provided needs to be correlated with the surgeon’s own knowledge of the surgical anatomy and the appropriate screw trajectories through that anatomy. Image-guided navigation is not a replacement for the surgeon knowing the pertinent spinal anatomy and surgical technique. It merely serves to help confirm a surgeon’s estimation of the nonexposed anatomy by providing image information that exceeds that provided by intraoperative fluoroscopy. Despite the advantages of image guidance, the surgeon must ultimately assess the information provided by these systems and determine whether it correlates with his or her estimation of the nonexposed anatomy and the proposed surgical plan. If sufficient correlation is present, the surgical step can be carried out. However, if sufficient correlation is not present, the surgeon needs to reassess both the spinal anatomy and the image-guided registration accuracy before proceeding.

Image-guided technology also has varying degrees of intraoperative functionality depending on the features of the navigational system used. This translates into an ease of use factor that can either simplify or complicate the overall procedure. Typically, the use of the surface mapping registration technique and a reference frame add time to the navigational procedure, frequently making it longer and more complicated than using fluoroscopy alone. The use of the paired point registration technique without a reference frame simplifies the spinal navigation process. Using this approach, the insertion of four pedicle screws typically takes no more than 8 to 10 minutes, and the need for standard fluoroscopy is eliminated for most spinal screw fixation procedures.

Fluoroscopic Navigation

Fluoroscopic navigation is the combination of standard fluoroscopy with image-guided navigational technology. It was developed to address the difficulties of some earlier image-guided systems that typically took much longer to use than standard fluoroscopy.²² Although standard fluoroscopy is employed with this technique, the amount of fluoroscopic time is significantly reduced.

With the patient in position before surgery, an anteroposterior and lateral fluoroscopic view of the pertinent spinal anatomy is obtained. This is done with a customized reference frame attached to the C arm. The frame serves to superimpose a specific reference grid on the two images obtained. The navigational work station can then take the two images with the superimposed grid and relate the spatial position of the imaged anatomy to a navigational probe. As the navigational probe is place on the patient’s anatomy, a corresponding trajectory line and cursor can then be superimposed on the lateral and anteroposterior images, respectively (Fig. 305-e12).

FIGURE 305-e12 Workstation screen of a fluoroscopic navigational system. Only the standard anteroposterior and lateral views are provided. Unlike computed tomography–based image-guided navigation, fluoroscopic navigation does not provide the critical axial plane view (selected trajectory highlighted by arrows).

Despite the advantages of fluoroscopic navigation, it still has some of the same difficulties that are experienced with standard fluoroscopy. Although the radiation dosage to the patient and surgical team is reduced, it is not eliminated. Positioning difficulties are the same in the upper thoracic region. Both the upper thoracic region and the lumbosacral region in obese individuals can be difficult to adequately visualize with fluoroscopic imaging.

The main disadvantage of fluoroscopic navigation compared with CT-based navigation is the image plane limitation. As with conventional fluoroscopy, fluoroscopic navigation provides the surgeon with only anteroposterior and lateral planar images. Unlike CT-based navigation, it does not provide an axial image, which in most spinal screw fixation procedures is the critical plane to identify intrusion into the spinal canal by a medially displaced screw.

A variation of conventional fluoroscopy, isocentric fluoroscopy, offers some improvements to the limitations of fluoroscopic navigation. This device acquires images intraoperatively by rotating the C arm in a 180-degree arc around the patient. As with conventional CT imaging, the acquired images can then be reconstructed into multiplanar images, including images in the axial plane. Although the images are not of the same quality as standard CT imaging, they are sufficient for navigational use. Image acquisition can also be repeated during surgery if needed to assess adequacy of decompression or screw positioning.

The most recent advancement in intraoperative imaging involves the use of a flat panel detector technology to improve intraoperative image acquisition and quality. A flat panel detector can be mounted onto a mobile imaging unit similar to a conventional C-arm fluoroscope. Although this unit can be used to acquire standard anteroposterior and lateral images, its C-arm configuration can be “closed” to completely encircle the patient. This allows the flat panel detector to be swept in a 360-degree arc around the patient, significantly improving the acquired image quality. The reformatted images are similar in quality to conventional CT imaging and superior to isocentric C-arm imaging. This ability to acquire high-quality planar images in addition to reformatted CT images makes this technology ideal for minimally invasive spinal surgery.

Conclusion

Image-guided navigational technology has been successfully applied to spinal surgery. It can be used for both conventional and minimally invasive spinal procedures. By linking digitized image data to spinal surface anatomy, image-guided spinal navigation facilitates the surgeon’s orientation to unexposed spinal structures, improving the precision and accuracy of the surgery. It is typically used to optimize the placement of spinal fixation screws and to monitor the extent of complex decompressive procedures. It can also be used as a preoperative planning tool.

Although image-guided spinal navigation is a versatile and effective technology, it is not a replacement for the surgeon having a thorough knowledge of the pertinent spinal anatomy as well as correct surgical techniques. It merely serves as an additional source of information used by the surgeon to make selected intraoperative decisions. In this way, it is similar to more conventional intraoperative imaging techniques (i.e., fluoroscopy), except that it provides a greater degree of image information to the surgeon.

Ideally, the clinical application of this technology to spinal surgery should facilitate a reduction in operative time, morbidity, and costs. It should be capable of minimizing or eliminating the need for conventional intraoperative imaging. It should be fast, easy to use, reliable, and capable of providing accurate intraoperative information while minimizing any disruption to the standard routine of each surgical procedure. Ultimately, it needs to be clinically versatile. It is the routine use of this technology by multiple surgical specialties that will drive its continued evolution and development as well as establishing it as a cost-effective surgical tool.