INTRODUCTION
Endoscopic spine surgery (ESS) is increasing in use and popularity, due to advancements in tools and techniques that allow for decompression of extrinsic neural compression with smaller incisions while allowing for direct visualization at the site of spine pathology [1-3]. Minimization of surgical dissection has been shown to result in reduced requirements for inpatient care, lower surgical expenditures, expedited recovery, and decreased postoperative pain [4-11]. Current spinal procedures using endoscopic visualization include both endoscopic decompression procedures in the cervical, thoracic, and lumbar spines as well as endoscopy-assisted surgery, which may involve both decompression and fusion procedures [1].
ENDOSCOPIC SPINE SURGERY AND TECHNIQUES
Both full endoscopic spine surgery (FESS), or uniportal procedures, and unilateral biportal endoscopy (UBE), or biportal procedures are currently being used for decompression surgery for stenosis as well as lumbar interbody fusion, with evidence to suggest that it is equally as effective as more traditional, open approaches [12-17]. Most studies discuss outcomes related to navigation in FESS rather than UBE, highlighting that navigation in UBE is still in the early stages of adoption (Table 1).
FESS consists of a singular working and endoscope channel and the utilization of a single instrument at a time. The single portal is relatively small and therefore irrigation with continuous saline is necessary to create space in between tissue planes. With FESS, surgeons have more limited operative exposures and decreased working areas in addition to an inability to use more than one instrument simultaneously and completely independently of the endoscope camera [2]. Image guidance can be used to assist in determining orientation, trajectory, and target during surgery. Intraoperative C-arm fluoroscopy, however, exposes the surgical team to significant amounts of radiation throughout the surgery [18]. Additionally, fluoroscopic x-rays project 2-dimensional (2D) imaging of the anatomy that is often of limited assistance. Therefore, various navigation and enabling technologies can be utilized for assistance with localization and assessment of the intraoperative target spinal pathology through the endoscope, including 3-dimensional (3D) C-arm systems, augmented reality (AR), and robotics [19].
UBE utilizes 2 portals for the endoscope and instrumentation, respectively. Advantages include independent scope and instruments that allow for concurrent visualization, retraction, cauterization, and bone resection. The surgical instruments are the same instruments used in traditional spine surgery such as Kerrison rongeurs, pituitary rongeurs, and curettes, which are familiar to spine surgeons. Similar to the FESS approach, continuous saline irrigation is utilized to create a temporary space. However, greater tissue dissection is required due to the need for multiple access portals. Navigation is not as commonly used in UBE, as determining anatomic orientation is simpler due to the use of various channels. Nonetheless, navigation can still be used as an adjunct for fusion, revision, and degenerative scoliosis cases to help determine anatomic orientation, with some surgeons selecting to use it for all UBE cases as well.
ESS has advanced in recent years and is now being commonly used for both decompression and fusion-related procedures. The limited surgical view however, can lead to disorientation through the endoscope resulting in increased risks and operative times. Therefore, the employment of navigation with the registration of patient-specific anatomy enables clear delineation of anatomic landmarks and facilitation of a successful procedure. We present a consolidated review of ESS and enabling navigation and emerging technologies to help clinicians gain a deeper understanding of the interplay between this rapidly growing surgical technique and the corresponding advancements in technology that allow for increasingly greater employment of ESS for more surgical indications.
IMAGE-GUIDED TECHNOLOGIES
Image-guided systems can provide real-time images to assist in surgical accuracy, stability, and precision in determining spinal anatomy and placement of implants endoscopically [20-24]. The placement of the skin incision and trajectory of the endoscope requires accuracy and precision, which may increase the use of image guidance [25]. Employment of this technology in ESS has been shown to provide useful and accurate information for appropriate placement of surgical tools through a 2D, endoscopic view [21,22,26-30]. There are multiple modalities available for surgeons to use, ranging from C-arm fluoroscopy, computed tomography (CT)-guided navigation, AR, and robotics.
1. Fluoroscopic Imaging
C-arm fluoroscopy has been the conventional modality used to confirm anatomic orientation and positioning of implants in both the lateral and anterior-to-posterior projections. Despite its resources and benefits, fluoroscopic imaging has the caveat of radiation exposure to the patient and the surgical team, the potential for added surgical time due to the specific nature of obtaining adequate and accurate views, as well as less precision and accuracy in implant positioning relative to navigation-enabling technology [27,30]. Additionally, fluoroscopy is often limited in providing accurate depth approximations or approach angles for instrumentation, requiring surgeon estimation of incision position and spatial orientation [31]. This creates numerous technical challenges for novice or less experienced surgeons who have not developed spatial and anatomic awareness. Lin et al. [32] reported on a retrospective, comparative analysis between CT-guided endoscopic surgery versus conventional fluoroscopic-guided for lumbar disc herniation with high-grade migration. Operative times, blood loss, and hospital length of stays were similar between the groups but the CT-guided cohort recovered faster and demonstrated decreased levels of postoperative pain immediately after surgery and at 1-year follow-up.
Two-dimensional navigation systems based on C-arm fluoroscopic images may also be utilized (Fig. 1). These modalities are less costly than 3D-based navigation with significantly reduced radiation exposure to the patient. However, the disadvantage of 2D navigation in endoscopic surgery is the lack of accuracy and detail of the axial view. On the other hand, the advantages of using 2D optical navigation include reduced procedure time and decreased radiation exposure to the patient, due to the reduced need for repeated fluoroscopic views, particularly during the transforaminal approach in endoscopic surgery (Fig. 2A and B).
2. CT-Guided Navigation
CT-guided navigation has become an increasingly popular adjunct to FESS, as it eliminates the need for multiple fluoroscopic radiographs and therefore reduces radiation exposure to the operating room staff [4,21,22,24]. Various systems are commonly utilized in spine surgery today including Airo CT, a intraoperative CT scanner mounted on the operating table and 3D C-arm systems such as O-arm, which is portable and can be moved from room to room. 3D C-arm systems are used to obtain a series of fluoroscopic images to create 3D CT-like reconstructions after the patient has been placed in the operative position and is under general anesthesia, and anatomical mapping can be performed with various tracking tools that can then be used for instrumentation and confirmation of anatomical positioning during ESS. In all these systems whether CT or CT-like, 3D reconstructions of the anatomy can be displayed on an intraoperative monitor and the operative tracking tools synchronized with the images in real-time. Surgeons and pioneers of ESS in Korea have been working on combining ESS and enabling technologies since the mid-1990s and began utilizing O-arm navigation systems for ESS when these systems were first installed in Korea in 2009 [33,34].
When utilizing CT-based navigation, there is increased radiation exposure to the patient, whether the CT scan is obtained preoperatively or intraoperatively. If preoperative CT scans are utilized, thin-cut CT imaging is required to improve the accuracy and precision of the system, which increases radiation exposure to the patient. This radiation exposure is cumulative with each radiograph and CT scan obtained perioperatively during the treatment course and lifetime of the patient. On the other hand, when employing 3D C-arm systems, the operating room (OR) staff may exit the OR during the intraoperative collection of the fluoroscopic images, reducing the radiation exposure to the staff and surgeon. Cumulative radiation exposure is a significant occupational risk given the numbers of cases during the careers of the surgeons and staff.
Navigation has been shown to assist with instrument accuracy and precise placement of hardware and devices during ESS, as these technologies provide 3D control of surgical instruments in real-time [21-23,27-30,35]. Additionally, given the challenges of orientation and complex interrelationship of anatomical structures during ESS, especially early in the learning curve or for a more anatomically complex patient, there may be a greater need to rely on fluoroscopic guidance. This may result in additional ionizing radiation from repeated use, increased time in the operating room, and a potentially elevated risk of vascular and neurologic injury [36]. Once the navigation workflow has been optimized, especially at high-volume centers, CT-guided navigation can lead to improved efficiencies in the procedures and has become an acceptable alternative to fluoroscopy with FESS at some centers. However, it does take time to achieve optimal navigation workflow for the OR staff and this can lead to increased surgical times for proper set up during this learning process, especially in centers that are not high-volume.
In the treatment of spondylolisthesis, FESS can be used in conjunction with navigation to allow surgeons to work in the central, lateral recess, and extraforaminal region as well as for percutaneous pedicle screw fixation [16,17,20]. Additionally, several authors have reported on the utility of CT-based navigation in ensuring accurate placement of the skin incision and trajectory and improving the learning curve, as well as facilitating successful navigation-guided endoscopic transforaminal lumbar discectomy, decompression, and interbody fusion procedures [37-44].
Historically, both FESS and UBE-TLIF procedures have been technically challenging due to the limited space for cage insertion, resulting in smaller cage footprints and theoretically increased risk for subsidence and pseudoarthrosis [45,46]. Intraoperative CT imaging guidance allows the navigation probe to be used to show the operating site and all anatomical planes in real time, including their relationship with the incision site, enhancing anatomical orientation. It can guide the spinal canal decompression, precise facet joint removal, disc space preparation, and cage positioning. It can optimize the procedure which enables the use of dual TLIF convergent cages, as reported by Suvithayasiri et al. [23] and also reduce unnecessary bone resection that could lead to further instability, as reported by Ho et al. [47].
Despite these advantages, there are several disadvantages of CT-based navigation including the placement of an additional incision for the pelvic pin that attaches to the reference array. This incision can be similar in size to the endoscopic incision itself and increases the invasiveness and pain of the procedure to some degree. In addition, any excessive movement of the patient by the surgical team, or accidental impact on the reference array can introduce errors in the navigation system and may require repeated scans to restore the accuracy of the navigation, which can lead to increased radiation exposure for the patient and additional time for the surgery. Multiple accuracy checks in distinct known anatomical landmarks are recommended to secure precision, which may impact the surgery workflow and increase surgical time. Accuracy is paramount, especially when placing the endoscope into the spinal canal risking dural tear and working closely to the exiting nerve root and dorsal root ganglion that also can be injured and result in temporary or permanent sequelae. In addition, CT-guided navigation may not be readily available at all hospitals, especially at ambulatory surgery centers where OR space is limited and cost considerations are a priority.
3. Electromagnetic-Based Navigation
Electromagnetic (EM) navigation relies on a control unit, an EM field generator, and a set of sensors [48]. Once the generator establishes an EM field, the sensors—attached to both the patient and surgical instruments—transmit their positional data back to the control unit, allowing the navigation system to track real-time instrument movement. The fundamental components of an EM navigation system involve the EM field interacting with sensor-equipped instruments to provide continuous spatial feedback.
Studies comparing EM navigation with conventional fluoroscopy (2D) in transforaminal endoscopic lumbar decompression found that the EM-based technique significantly reduced both total surgical time and the time required to place the cannula [49]. Furthermore, it lowered overall radiation exposure without compromising functional outcomes when compared to fluoroscopic methods [50]. Similar findings have been reported for pedicle screw placement, in which EM navigation was associated with shorter operations, reduced radiation, and improved accuracy. Part of the efficiency gains may stem from more precise guide wire and working tube placement within the safe triangle of the spinal canal [51,52]. Additionally, EM navigation facilitates the learning of minimally invasive spine surgery techniques using various instruments, including endoscopes, puncture needles, and catheters [49,50].
Xu et al. [20] reported on a comparison between an optical (3D) navigation system and an EM-based navigation system for use in FESS transforaminal lumbar interbody fusion, with results suggesting that the 2 techniques yield similar surgical outcomes. Koivukangas et al. [53] demonstrated that EM-based navigation systems had slightly lower accuracy than optical systems, with the technical accuracy of EM systems noted to be 0.30± 0.13 mm versus 0.20± 0.10 mm for the optical system. This difference, however, is noted to be of limited clinical significance. Optical navigation systems rely on infrared light-emitting diode (LED) light tracking, captured by receiving cameras that track based on the movement of LED light and integrated into patient-specific anatomy obtained via preoperative imaging (Fig. 3A-G). These systems require dynamic reference bases and a continuous line of sight to the signal emitters, which may result in impaired navigation if the line of sight is disrupted [24,54-56]. A significant advantage of optical navigation systems is their universal adaptability to various surgical instruments through simple reference frame attachment. This versatility allows surgeons to track multiple instruments simultaneously by simply securing tracking arrays to any surgical tool, making it highly flexible during different phases of the procedure. EM navigation, on the other hand, relies upon internal reference electrodes contained within the surgical instruments and EM fields that course through the patient’s body and is significantly less bulky than the optical navigation systems [24,55,57-60]. Additionally, these systems are devoid of the line-of-sight limitations that are encountered with the optical navigation systems [20,24,54,55].
Despite these advantages, current evidence remains insufficient to definitively establish EM navigation as superior to nonnavigated approaches in ESS. More extensive research should address a range of parameters—including radiation exposure, learning curves, cost, and clinical outcomes—to provide a clearer understanding of its comparative benefits and limitations. Nonetheless, for better understanding the difference between all types of navigation-assisted in ESS, a summary of key pros and cons between fluoroscopic (2D), optical (3D), and the EM-based is depicted in Table 2.
4. Augmented Reality
AR involves computer-generated, 3D reconstructions of patient-specific anatomy that overlays over the patient’s anatomy seen intraoperatively through the AR headsets, allowing surgeons to assess real-time, intraoperative trajectories, anatomical relationships and improve localization with potentially improved ergonomics (Fig. 4) [24,61,62]. AR systems are an extension of CT navigation as the 3D reconstruction images are similar to those displayed on the OR monitors but displayed through the headset closer to the surgeon’s eyes (Fig. 5). This improves the ergonomics and neck position of the surgeon. Some systems can project different imaging exams and allow surgeons to interact and scroll through the sequences by hand gestures. Pierzchajlo et al. [63] review available AR technology in minimally invasive spinal surgery, describing the most prominent AR systems as well as their utility in surgical training and for pedicle screw placement models. Jamshidi et al. [64] report on a single case of AR use in full endoscopic transforaminal interbody fusion, citing advantages such as smaller incisions and tissue preservation, reduced pain postoperatively, and improved safety and precision. These results should be interpreted with caution, however, given it reflects a single case experience. Overall, while there is a dearth of literature reporting on the concomitant use of AR and FESS, Hasan et al. [61] believe there is potential for AR to not only significantly reduce the learning curve of ESS but also assist in surgical execution. At this stage however, current AR systems are in their infancy and do not incorporate endoscopic video with the 3D-CT imaging and navigation into the AR headsets for direct view and intraoperative use by the surgeon. At this time, AR can only be utilized in the initial portion of ESS cases with portal planning and docking of the endoscope, or the percutaneous pedicle screw placement in endoscopic fusions since the headsets would be removed to visualize the endoscopic video on the OR monitors. The headset weight would also be a consideration as the headset may cause discomfort or disorientation with prolonged use.
5. Robotics
Robotic technology has been used as an adjunct in full ESS to help with pedicle screw placement, determining orientation and trajectory for endoscopic discectomy, and percutaneous interbody delivery, with authors citing synergistic use of these technologies to perform the procedure and identify percutaneous trajectories through Kambin triangle [24,65]. The trajectory of the robotic arm can be planned in advance through preoperative CT or intraoperatively with a 3D C-arm and is an extension of navigation technology. The robotic arm can be rigidly held in place to improve the accuracy and placement of the endoscope into the target anatomical region. Additionally, Li et al. [66] report on the utilization of robotic assistance for endoscopic laminotomy procedures, with the robotic arm being used for planning the skin incisions, targeting of the interlaminar window, execution of precision drilling, and depth control. Recent literature has also reported on the safety and effectiveness of using a robot for tasks other than pedicle screw placement, including accessing Kambin triangle in percutaneous interbody fusion procedures [67]. This evidence suggests that robotic technology in ESS is a promising assistive tool that may continue to become more prevalent in practice as robotic technology becomes more commonly utilized. However, the adoption of robotics has thus far been slow mainly due to the high cost of these robotic systems.
AVOIDING COMPLICATIONS IN ESS WITH THE USE OF NAVIGATION
Although enabling navigation technology in ESS promises to improve accuracy and safety, complications may occur. One of the most significant concerns when performing ESS with the use of navigation is an alteration of reference point locations and inaccurate mapping and registration for stereotactic imaging. Inaccuracies often arise from the reference frame, the patient’s body movement, or the instruments used for navigation. Ensuring the following steps in patient positioning and setup are optimized may help mitigate this risk:
1. Patient and Reference Frame Positioning
(1) Ensure that the patient and operating table are optimized to account for lordosis and maintain ergonomic positioning, particularly for the L5–S1 level. To prevent drifting up to L4–5, position the table with the head elevated and consider adding more reverse Trendelenburg.
(2) Ensure the position of the reference pin is secure (Fig. 6).
(3) Confirm that the reflective marker spheres are accurately positioned.
(4) Verify that the navigation screen displays that all spheres are simultaneously in the system’s range of view, indicating good reflection from the marker spheres.
2. Patient Body Movement (Vertebral Body)
(1) Ensure there is no relative movement of the vertebral body to the reference frame.
(2) Avoid rotation and translation of the patient’s body.
(3) In endo-fusion cases, be aware that navigation may become inaccurate after cage or screw insertion.
3. Instrument
(1) Ensure there is no bending of the navigation instrument.
(2) Use the navigation probe and dilator for guidance during endoscopy.
The use of intraoperative CT-based navigation can be especially beneficial in academic settings for trainees or during the learning phase for surgeons. It can also facilitate procedures and enhance safety, even for experienced surgeons. By providing instant visual reinforcement of what is seen through the scope and comparing it to CT images, trainees can more easily learn the anatomical landmarks and steps involved in performing surgery safely.
For instance, one technique commonly used in revision cases or transforaminal approaches involves a custom dilator (Fig. 7A-C). The third dilator, which is usually the last dilator in most endoscopic sets, is cut in half. This dilator has an inner diameter large enough to accommodate the entire shaft of the navigation pointer, allowing it to be used during the initial approach. Before introducing the pointer to the patient, the custom dilator is coupled with the pointer. The pointer can then be easily docked to any anatomical landmarks using navigation, reducing biplanar fluoroscopic usage. Once the pointer is docked, the dilator is pushed down to the landing spot, and the pointer is removed. This technique eliminates the need to switch between the pointer and the initial dilator, as the pointer functions as the initial dilator. This approach helps ensure that the targeted landing spot is not lost once it is found.
CURRENT LIMITATIONS OF IMAGE-GUIDED TECHNOLOGY
FESS offers the option for more minimally invasive surgery that can achieve the same surgical objectives as more conventional surgery with significantly less collateral tissue damage. Inherent limitations involve difficulty obtaining optimal visualization and access to the spine, as well as an anatomic complexity that may result in increased risks to the patient. Therefore, concomitant use of image-guided navigation may be beneficial for surgical planning and execution in FESS, with emerging data supporting its use.
However, there are distinct disadvantages to the use of emerging navigation technologies with ESS including increased cumulative radiation exposure to patients with each CT or 3D Carm spin performed, the early state of technology with current AR systems, and the high cost of navigation and robotic systems that may not align with financial reimbursement of simple endoscopic discectomy or decompression. In the current state, these emerging technologies may be better utilized in multilevel fusion surgeries for more accurate, precise, and efficient placement of implants, especially as the number of implants and surgical levels increase. As the technologies advance over time with reduced radiation exposure to the patient, more advanced AR systems combining high definition endoscopic or even 3D video with 3D real-time navigation imaging through the AR headset, and significantly reduced cost for these advanced technologies, the merging of ESS and these emerging technologies may become more commonplace.
CONCLUSION
ESS encompasses both FESS and UBE approaches, which provide access to a variety of lumbar spinal pathologies while minimizing damage to the surrounding soft tissues. Image-guided systems, such as C-arm 2D fluoroscopy, CT-guided navigation, EM navigation, AR, and robotic technology, enhance surgical accuracy and precision in identifying spinal anatomy and placing implants during ESS. Although there is a learning curve associated with adopting these techniques, proper training, careful planning, and meticulous execution can lead to excellent outcomes.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Print
