31 Neuronavigation for Complex Thoracic Spine Surgery

The advent of spinal instrumentation and the development and use of screw-based fixation devices greatly influenced the management of spinal disorders, allowing for treatment of more complex spinal pathology while maintaining or restoring spinal stability and alignment.^1,2,3

As spinal procedures progressed and became more complex, it became crucial to minimize the injuries associated with incorrectly positioned implants and screws that, when misplaced can cause spinal, nerve roots and vascular injuries as well as dural tears with cerebrospinal fluid leakage.^2,3

Pedicle screw-based instrumentation remains one of the strongest posterior fixation techniques for the thoracolumbar spine and it is the standard procedure for treatment of thoracic spine disease. By traversing all three columns of the vertebrae, pedicle screws can rigidly stabilize both the ventral and dorsal aspects of the spine and the rigidity of pedicle fixation allows for the incorporation of fewer normal motion segments to achieve stabilization of an abnormal level. Furthermore, pedicle screw fixation does not require intact dorsal elements and so it can be used after a laminectomy or traumatic disruption of laminae, spinous processes and/or facets.⁴

However, in comparison with the lumbar spine, in the thoracic spine there is admittedly a much lower margin of error, as errant screws are capable of injuring the spinal cord and other structures intimately related to the vertebrae, including the thoracic pleura, esophagus and intercostal and segmental vessels. Other structures within the thoracic cavity at potential risk include the thoracic duct, azygous vein, inferior vena cava, and aorta.

Moreover, wide variations in diameters and angles of thoracic pedicles, depending on the thoracic level, have been documented in anatomical studies. Therefore, placement of thoracic pedicle screws can be even more challenging, as pedicle angles and attachment to the vertebrae tends to be more anatomically varied in the thoracic spine, especially at the middle thoracic levels, which have the narrowest pedicles and a closer proximity between the medial pedicle wall and the spinal cord. Minor deviations in the starting point or angulation can result in marked malpositioning of the screw.⁵ The smallest pedicle is typically found at T4 and its diameter can be as small as 4.5 mm and the largest pedicle is usually T11 or T12, with about 8 mm diameter.^6,7 The risk of malpositioned screws is even higher in spinal deformities where vertebral anatomy can vary even more widely.^8,9

The classical technique for pedicle screw insertion is the “free-hand” pedicle screw insertion and is essentially a blind technique where adequate screw placement depends on correct identification of anatomical landmarks, surgeon experience, and reproducible technique.¹⁰

As such, early on, the learning curve associated with usage of this technique became apparent, leading to increased surgeon usage of image-assisted techniques. The technology used to acquire imaging for intraoperative navigation has evolved from the discovery of X-rays in the late 19th century to the highly sophisticated intraoperative computed tomography (CT)-based navigation tools used today.^3,11,12 Before the advent of spinal image guidance, surgeons relied on their knowledge of anatomy, complemented with the preoperatively image examinations and intraoperative imaging such as serial radiographs and later on fluoroscopy. Although plain radiography is still used by some surgeons to assist localizing the skin incision, determining the proper anatomical level, and confirming satisfactory position of the spinal implants prior to closure. Unfortunately, conventional radiography processing is time consuming and only static images can be acquired. Therefore, instantaneous positional information regarding instrument position within the surgical field cannot be obtained.

Fluoroscopy, namely C-arm fluoroscopy, addressed some of the disadvantages of plain radiography. Hence, spine surgeons started to use this tool as their primary mean of intraoperative navigation. This technique allows real-time imaging when used continuously (e.g., to obtain immediate updates of an instrument’s position), or it can be used to acquire multiple static images in succession.¹³ One of its main shortcomings is the potential for significant occupational radiation exposure, especially when continuous fluoroscopy is used.^14,15 The surgeon is at particular risk due to their close proximity to the fluoroscope and the patterns of radiation scatter. Also, when using a single fluoroscope, images can only be obtained in a single plane at a time. When simultaneous biplanar fluoroscopy is needed, two independent C-arms are required for intraoperative navigation, which can create some create ergonomic constraints that hinder access to the surgical field and potential for breeches in sterility.¹³ Furthermore, this technique provides only two-dimensional imaging of complex three-dimensional structures, leaving to the surgeon extrapolate the third dimension based on his/her interpretation of the images and the knowledge of pertinent surgical anatomy.¹² This conventional intraoperative imaging cannot provide the axial plane, which is for most screw procedures, a critical plane to confirm precise screw placement.¹² Fluoroscopy is used so often during pedicle screw placement that it has been referred to as the “conventional” method, perhaps reflecting its almost expected usage when attempting to employ free-hand techniques.

The conventional techniques used for pedicle screw insertion are based on anatomical landmarks either without image guidance (“free-hand technique”) or with fluoroscopic guidance in a lateral, anterior–posterior, or oblique ventrodorsal projection.¹⁶

In order to help the surgeon obtain optimal thoracic screw placement, different entry points, screws trajectories, and insertion techniques have been described.^{7,8,17,18,19,20,21,22} However, those are not discussed in this chapter, as it focuses on the landmarks relevant to applying navigation techniques.

In vivo and in vitro studies, in which free-hand technique and fluoroscopy-guided techniques have been used, report thoracic pedicle misplacement rates, from 3 to 55%.^8,23 Even experienced surgeons misdirect the screws medially in 5% and inferolaterally in 15% of the cases, using fluoroscopic imaging.⁵

The accuracy is achieved by surgeon’s expertise and familiarity with the surgical anatomy and is greatly facilitated through intraoperative imaging. This is especially pertinent when the surgeon needs to be precisely oriented to that part of the spinal anatomy that is not exposed in the surgical field. This is the case when using minimally invasive spinal surgery (MISS) procedures, when the visualization based on anatomical reference points that can be used as a basis for orientation and implant placement is not available.^2,12,24

Computer-assisted spine surgery (CAS) is a computer-based technology that links spine image, acquired by conventional techniques, such as fluoroscopy and CT, with an accurate representation of intraoperative anatomy.^25,26 This gives the surgeon the ability to manipulate multiplanar CT or fluoroscopic images during the surgical procedure in order to gain a greater degree of orientation to the nonvisualized spinal anatomy and, therefore enhancing the accuracy of spine surgery, of upmost importance during spinal instrumentations. It can also minimize the surgical team’s exposure to radiation.

Navigation has been questioned in terms of accuracy, namely concerning pedicle screw insertion, and it has shown to attain high accuracy (> 95–99%)^{23,27,28,29,30,31,32,33,34,35,36} in comparison with free-hand and/or fluoroscopy-based techniques and a lower rate of screw-related complications.^29,30 Especially the latest navigation technology set-ups seem to have virtually eliminated the need for reoperation for screw malposition, as they allow for intraoperative confirmation of implant positioning.³⁷ Furthermore, the navigation techniques allows for decreased radiation exposure for the surgical team and operating room (OR) staff.^38,39

31.1.1 Historical Perspective on Image-guided Spinal Navigation

The image-guided spinal navigation, or CAS, evolved from the principles of stereotaxic, which consists in localizing a specific point in space through the use of three-dimensional coordinates combined with optoelectronic position sensing techniques.^2,11,40 The technology behind this was initially developed for intracranial neurosurgical procedures. At the beginning, stereotaxic demanded the use of an external frame attached to the patient’s head and even so, the procedures were associated with a certain degree of inaccuracy, mainly due to the movement of brain structures during retraction and resection of brain tissue (brain shift).¹¹ Applying the same principles to spinal surgeries was more challenging both because an external frame was unpractical and of the lack of anatomical constancy in a moving spine. Furthermore, the skin and underlying soft tissue are mobile relative to the spinal column and it was therefore necessary to use bony landmarks for registration, what requires an extensive and meticulous surgical exposure.⁴¹

Even though the use of intraoperative navigation was not initially compatible with spinal anatomy, there was great demand for this technology among spine surgeons who felt that navigation would be especially useful in situations where spinal implants were placed without direct visualization, such as placement of pedicle screws.⁴¹

With the evolution of computer-based technologies in the 1990s appeared the frameless navigation technology, raising the possibility of using stereotaxic on other different procedures, namely, spinal navigation.¹¹ Brodwater and Roberts, in 1993,⁴² published the first attempted transition from intracranial to spinal surgery, using an image-guided microscope and skin surface fiducial markers for registration. However, these markers were subject to movement in reference to the underlying spinal anatomy, resulting in significant navigation inaccuracy.¹²

Kalfas et al^25,26 and Nolte et al^40,43 demonstrated feasibility of using navigational technology to improve accuracy of lumbar pedicle screws insertion and Foley et al⁴⁴ described the use of easily distinguishable anatomical landmarks on the posterior aspect of the spine as fiducial points, in conjunction with a dynamic reference array that was fixed to the spine, partially solving the issue of inaccuracy. With the evolution of spinal navigation, the necessity of intraoperative imaging became increasingly important to provide accurate registration of the spine to the navigation system. And, the most recent intraoperative CT-based image-guided (CT) navigation has improved, including localization, real-time navigation, phantom trajectory prediction, and better definition of bony and soft tissues.^28,45,46

31.1.2 Principals of Image-guided Spinal Navigation

The initial step in integrating CAS with any type of spinal procedure is the acquisition of multiple successive images of the region of interest, a process that may be accomplished with either fluoroscopy or CT.⁴⁷ The CAS or image guidance technology is available in a variety of setups which may be differentiated according to the manner in which these images are captured, processed, and presented to the surgeon.⁴⁷ Usually, the elements that compose a navigation system are: (1) a system for image acquisition that allows tracking of specialized instruments in relation to a single or multiple reference points attached to suitable anatomical landmarks and (2) a computer workstation that reconfigures this data set into a series of multiplanar images that are displayed on a monitor along with the relative position of any instrumentation within the operative field.⁴⁷

Image Acquisition

Usually, the elements that allow image acquisition consist of a two-camera optical localizer that interfaces with the image-processing computer workstation through emission of infrared light to the operative field or an electromagnetic registration system. Passive reflective spheres placed in a hand-held navigation tool serve as the connection between the surgeon and the computer workstation. These passive reflectors can also be attached to the traditional surgical instruments like the drill guide, a tap, or a pedicle screwdriver. In order to accurately calculate the position of the instruments in the surgical field and the anatomical points, where the tip of the instruments is resting, the spacing and positioning of the passive reflectors on each navigational probe or customized trackable surgical instrument are programmed into the computer workstation. In fact, after the infrared light is transmitted toward the operative field and is reflected to the optical localizer by the passive reflectors, the information is conveyed to the computer workstation, allowing the calculation of the spatial location by matching spinal image data (CT or fluoroscopic images) to its corresponding surgical anatomy.^2,3

Registration System

The accurate translation of spatial information into detailed renderings of spinal anatomy necessitates a stable frame of reference that enables the computer-assisted system to calculate the relative positioning of instruments within the surgical field in all three dimensions. This process of establishing a relation between the “real” coordinate system, as defined by the patient’s array, and the “virtual” coordinate system of the imaging data is called registration. Different registration techniques can be applied.^11,12

Point Matching Registration Technique

Several anatomical points are selected in CT and MRI data set and in the corresponding anatomy. These points have to be selected for each spinal level to be instrumented. Any anatomical landmark that can be identified both preoperatively and intraoperatively can be used as a reference point. Examples of these frequent points are the tip of a spinous or transverse process or the apex of a facet joint. After selecting one of these points in the CT image data, the tip of the navigation tool is placed onto the corresponding point in the surgical field, with reflective spheres on the tool handle aimed toward the camera. Infrared light from the camera is reflected off the spheres toward the camera and into the computer, which calculates the spatial position of the probe’s tip and the anatomical structure it is resting on. This effectively “links” the point selected in the image data with the point selected in the surgical field. If a minimum of three points are registered, when the probe is placed on any other point in the surgical field, the corresponding point in the image data set will be identified on the computer workstation. The disadvantage of this technique is that any error on the part of the surgeon in selecting the specific anatomical point in the surgical field will result in varying degrees of navigational inaccuracy. This paired point protocol can also be performed in conjunction with surface matching.¹²

Another method of registration is CT–fluoroscopy matching. This technique is used sometimes, when the navigation system is preoperatively CT based. When using this technique, the preoperative CT is matched with intraoperative two-dimensional fluoroscopy images of the spine, which are taken from different angles of the patient.⁴⁸

Surface Matching Registration Technique

Surface matching is a supplementary registration technique in which the surgeon randomly selects multiple anatomical points on the exposed posterior elements to provide supplementary topographic data. This technique does not preclude prior selection of the points in the image set, although several discrete points in both the image data set and the surgical field are frequently needed to improve the accuracy of surface mapping. The positional information of these points is transferred to the workstation, and a topographic map of the selected anatomy is created and “matched” to the patient’s image set. The concomitant use of the point-to-point and surface matching approaches was shown to result in a significantly lower mean registration error compared with that of paired point matching alone for the insertion of pedicle screws.¹²

Automated Registration

Automated registration is performed without any input from the surgeon and with less potential for registration error. It can be performed only when the image data are acquired intraoperatively. This technique involves attaching a reference frame with reflective spheres to some site in the exposed spinal anatomy or, in lumbar surgery, to the iliac crest. A second reference frame is built into the intraoperative CT imaging scanner or fluoroscope. As the intraoperative images are acquired, the two reference frames allow registration to occur without the need for the surgeon’s input. The CT scanner or fluoroscope can then be removed, and real-time navigation of up to five separate spinal levels is performed.^2,47,49

Tracking System

The image guided navigation systems use either optical or electromagnetic tracking systems.⁴⁷

Optical Tracking

These systems have been developed as another method for tracking the location of instruments during surgical navigation to address the disadvantages of optical devices, namely the need of a clear “line of sight” between the tracking device and the passive signal emitters (reflective spheres) in the surgical field.⁵⁰ This may restrict the operator’s normal range of movement and thus limit the intuitive handling of the instruments. Moreover, the trackers needed for optical systems with active and passive reflectors are attached to the instruments and the operation areas in order to be referenced and have anatomical and ergonomic disadvantages. Also, the instruments used are significantly larger and heavier, resulting in poorer ergonomics and handling for the operator.

In the electromagnetic registration systems, three orthogonal electromagnetic fields are generated by a transmitter attached to a fixed anatomical reference point, such as a spinous process. The positional data of these instruments are collected by a receiver and integrated to facilitate navigation. Since a line of sight is not required, the surgeon and the nursing staff are able to work freely within the operative field. However, electromagnetic registration image guidance may be compromised by metal artifacts, including surgical implants, as well as by any electromagnetic fields originating from other equipment in the OR, such as monopolar electrocautery, electrocardiogram monitoring, and cell phones. Given the limited area of these electromagnetic registration fields, the transmitter may also need to be repeatedly transferred to additional anatomical structures to obtain sufficient tracking information for multilevel procedures.^47,51

31.1.3 Types of Navigational Systems

As highlighted in Fig. 31.1, the two main types of navigation systems are intraoperative-based image and preoperative-based image.

Fig. 31.1 A diagram showing the differences between intraoperative and preoperative navigation systems.

The two primary options for intraoperative imaging of the spine are still radiography and fluoroscopy. C-arm fluoroscopy remains a low cost and widely available mode of intraoperative image acquisition and allows for rapid and serial visualization of two-dimensional images in real time.¹³

Nonetheless, the past three decades of image-guided spine surgery have witnessed the development of multiple modalities for intraoperative imaging and navigation. The ultimate utility of these technologies depends on critical appraisal of the unique advantages and disadvantages of each system.^12,41

Computer-Based Tomography/Preoperative (CT)-Based Navigation

The first available mode of intraoperative navigation was this modality. This technique uses preoperative thin-slice scans and one of the several registration processes to create a data set, which forms the basis for intraoperative navigation. In fact, prior to the surgery, a two-dimensional thin-cut CT through the region of interest that is obtained and uploaded to a workstation where it creates a virtual three-dimensional reconstruction that can be used for planning the surgery simulates the implants. On this preoperative reconstruction, anatomical landmarks are selected for intraoperative registration.^41,48

However, preoperative CT scans are acquired with the patient in a supine position, while during surgery patients are usually in the prone position. The resulting vertebral shift and realignment create a risk for navigation errors. Therefore, to account for shifting anatomy during surgery, each level must be registered separately to accurately plan and perform the surgery.^13,52 One significant disadvantage of CT-based guidance systems is the need for surgeon-dependent registration of anatomical landmarks on preoperative CT images and the corresponding anatomy of the patient intraoperatively. In addition, extensive bony exposure is required for adequate registration and it may be difficult to identify landmarks for registration in patients with prior laminectomies.^13,41

During navigation, the surgeon is presented with reformatted CT images on virtual three-dimensional multiplanar image reconstructions, along with the selected screw entry point and trajectory superimposed on the images. This information updates in real time, as adjustments are made to the selected trajectory in the surgical field.⁵³

The disadvantages of this technology includes increased radiation exposure to the patient preoperatively.¹³

Intraoperative Image-Based Navigation

Intraoperative-based navigation systems eliminate the need for surgeon-dependent registration step because the system is automatically registered during the acquisition of images intraoperatively. Thus, the need for spinal exposure for point matching is obviated. Furthermore, in this setting, as images are obtained after patient positioning, they are an accurate representation of vertebral anatomy at the time of surgery.⁴⁹

Two-Dimensional Fluoroscopy-Based Navigation

“Virtual fluoroscopy” or two-dimensional fluoroscopy-based navigation is a strategy that combines a standard two-dimensional C-arm with a computer navigation system. It uses a standard anterior–posterior and lateral image of the spinal anatomy acquired immediately before the start of the procedure.⁵³ Registration is performed automatically with a reference frame attached to the C-arm.⁴¹ A series of fluoroscopy images in anterior–posterior, lateral, and sometimes the pedicle oblique view are acquired with a reference frame attached to a stable anatomical landmark, often a spinous process in the vicinity of the vertebrae that will be operated. These images are transferred to the navigation workstation and this data set is used to navigate implants on the virtual anatomy viewed on the screen. An infrared camera aimed at the reference arc and navigation tools allows continuous recognition of the navigation tools in relation to the relevant anatomy. A continuous “line of sight” must be kept among the infrared camera, the reference arc, and the navigation tools. The accuracy of the system will be maintained as long as the stability of the reference arc is maintained, motion segments do not change their position compared to acquired images, and the navigation tools are kept in line with the desired trajectory. Therefore, fluoroscopy-based navigation allows a completely automatic registration of the spine, correction of image distortion, and reduction of radiation exposure to the staff, yet is limited to two-dimensional projection images. The risk of navigation errors is increased and the abnormal axial anatomy is more likely to remain unrecognized, as it does not provide three-dimensional visualization of the spinal anatomy during navigation. Errors may also be greater in cases of poor bone quality, excess intra-abdominal gas, morbid obesity, spinal deformity, prior surgery, and congenital anomalies. Furthermore, image resolution is typically best in the center of the field and any structures around the periphery may appear distorted secondary to parallax, so to maintain the accuracy of navigation across several spinal segments the process of data acquisition and anatomical registration may need to be repeated several times.^13,41,49

In summary, two-dimensional navigation provides an easier learning curve for surgeons who are familiar and very comfortable with conventional anterior–posterior–lateral fluoroscopy. It is readily available and is easily incorporated into the workflow when using devices or techniques that require real-time X-ray control and fluoroscopy to be brought into surgery. Such procedures include cement injection in vertebroplasty procedures, kyphoplasty procedures, or while using K-wires. Two-dimensional navigation can also be applied in situations that require frequent radiographic updates because of anatomy changes during surgery, such as translumbar discectomy and fusion procedures like extraforaminal lumbar interbody fusion (ELIF) where the cage placement and its relationship between vertebrae requires X-ray exposure. In these situations, two-dimensional navigation (NAV) will not eliminate the need for fluoroscopy, but can significantly reduce the need of continuous X-ray and will help improve the workflow. For example, two-dimensional NAV reduces the operative time and the risk of intraoperative contamination because the C-arm does not need to be continually repositioned during the case. This strategy is also preferable to preoperative CT-based navigation, which necessitates a suitable preoperative CT that must be matched to the corresponding anatomical structures by a manual registration process before proceeding with navigation.

This navigational system is a relatively inexpensive system (compared to three-dimensional navigation) that combines the benefits of navigation with affordable purchasing and maintenance costs, which can be of utmost importance when there are economic and infrastructural challenges. We have encouraged the use of navigation technologies whenever possible. We advocate the use of navigation whenever possible. During our annual mission in Tanzania, we effectively helped implement potable hardware and two-dimensional navigation system in a hospital that had no prior experiences of using such equipment ( Fig. 31.2).⁵⁴

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

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Clinical Presentation of Thoracic Spinal Compression
2. Proximal Junctional Deformity
3. Nonoperative Neurological Diseases of the Thoracic Spinal Cord
4. Intraoperative Neuromonitoring During Thoracic Spine and Spinal Cord Surgery

Stay updated, free articles. Join our Telegram channel

Join