The basic principle of modern neurosurgery is precise lesion localization that results in a minimally invasive approach.1 To achieve this goal, tools for localization called neuronavigation systems have been developed (Table 1) and are considered standard in today’s neurosurgical armamentarium. In nautical terms, the Latin verb navigare means “to steer a ship.” As in maritime or automotive navigation, neuronavigation requires maps (radiological images) of the areas to approach, localization tools (tracking devices) for one’s current position, and a mathematical function for the relation between real position and position on the map (registration).2,3
Registration, the key step in navigation-guided surgery, is defined as the determination of a geometrical transformation that relates the radiographic images to the physical patient.4 In commercially available navigation systems, rigid transformation of the skull consisting of translational and rotational movements is applied, and tissue deformations are not considered. Registration can be performed by matching sets of paired points or by aligning surfaces.
The paired-point registration is based on 2 corresponding sets of points on the patient’s head and on the image volume that are matched by the system to minimize their distance. As an estimation of the system’s accuracy, navigation systems usually display the root mean square error (RMSE) of the distance between corresponding points after registration. This registration error does not always correlate with the target error in both theory and clinical application.5,6 However, the automatically calculated registration error can be regarded as an indicator of whether the navigation system is working properly in a given case.4
Different types of points on the patient’s surface have been used for navigation registration. Bone screws are the most accurate markers but are rarely used because of their invasive application.7,8 Radiopaque and magnetic resonance (MR)-visible self-adhesive skin fiducials are the most widely used markers for registration today.9 Although easy to use, the application of skin markers is time-consuming, requires trained personnel, and causes patient discomfort because hair may need to be shaved. There is also a slight degree of patient risk because additional imaging is necessary. Furthermore, skin fiducials can shift over time, and registration accuracy is inversely related to the time span between imaging and surgery.10 Because of these drawbacks, alternative markers for registration have been sought. For instance, paired-point registration with anatomic landmarks9-11 provides the advantage of reducing staff and imaging time but needs experience for identification of corresponding points on radiographic images and patient anatomy. Furthermore, accuracy of landmarks registration is lower than with skin fiducials.9,10
Surface-based registration matches surface points acquired from the physical patient’s anatomy with a surface from the image volume. The “fitting the hat on the head” problem of 2 surfaces originally described by Pelizzari and Chen12 in 1987 is implemented in commercially available navigation systems as some adapted version of the “iterative closest point algorithm,”13 which searches for a transformation matrix of the minimum distance between the 2 surfaces. Usually, at least 200 points on the patient’s craniofacial surface are collected by direct probe contact or indirect laser14-16 or by 3-dimensional surface scan.17 Surface-based registration obviates the need for skin markers; the method is fast, intuitive, and easy to use; does not require the identification of corresponding points on patient and images; and therefore does not require the high level of experience of anatomic landmark registration. Registration can be performed with preoperative scans and even with scans that are limited to parts of the skull. However, errors may occur as a result of local relative minima; ie the iterative algorithm may stop if the matching reaches a local minimum that is not necessarily the best overall matching.3 In cases of laser scanning, which uses mainly the facial anatomy for point collection, surface-based registration has been found to be highly accurate in frontal areas but lower in occipital areas.16,17 In our experience, however, this does not occur with direct probe-contact scanning. Using a hybrid method of paired-point registration with anatomic landmarks and surface-based registration with 40 surface points, Pfisterer et al9 reported high accuracy corresponding to fiducial marker registration.
In today’s routine clinical setting, additional radiographic scanning for skin marker registration is unacceptable because of patient exposure to radiation (computed tomography [CT] scan) and time constraints (MR imaging [MRI]). From our experience with > 1500 navigation cases since 1997, we propose a surface-based (or hybrid) registration method for the clinical routine.
The primary principle of a navigation system is its localizing technology. This tracking device is used to determine the spatial position of a tracked instrument relative to the patient (via a tracker attached to the head) during the intervention. Different technologies of such position-sensing devices have been developed over time; whereas mechanical arms with potentiometer joints18-21 and ultrasonic systems22 are of historical interest (see above), currently available techniques include optical and electromagnetic (EM) systems (Table 1).
Optical systems use a digitizing camera system to localize trackers in space. The camera array detects infrared light beams either from LEDs (light-emitting diodes) in active systems, or from infrared light emitted by LEDs around the camera lenses and reflected from spheres on the patient and instrument trackers in passive systems. Because the geometry of the tracker is known, the position of the tracked instrument tip relative to the patient reference tracker can be calculated by the system using triangulation.
Although optic navigation represents a longstanding and proven technology that provides high submillimetric accuracy and a large tracking volume of several cubic meters, its biggest disadvantage is the necessity of a free line of sight between the camera and tracker because at least 3 infrared light-emitting/reflecting points must be visible on the trackers for successful localization. In cluttered operating setup environments such as microneurosurgery, however, the view between the tracking camera and the surgical site is commonly blocked by surgical tools (operating microscope or endoscope), eccentric patient positioning/approach, or a crowded operating field.23,24 The reference frame attached to the patient’s head is bulky and may partly obstruct the surgeon’s approach. Furthermore, optic tracking systems rely on a rigid geometric configuration between the tracking points and the instrument tip. Optical systems are therefore not suitable for tracking flexible devices such as catheters, needles, or flexible instruments.23
EM tracking technology is based on a magnetic field emitted by a generator as a coordinate system and coils in tracking devices that are brought into this field. The current induced in those coils is resolved for position detection. Besides a magnetic field generator, an EM navigation system consists—just like optic navigation systems—of a patient reference tracker fixed to the patient’s head and a localization tracker that is a probe for current position identification. Multiple coils in the localization tracker are used for tip and trajectory (2 coils) and rotation (3 coils) detection.
Although Kato et al25 developed an EM computer-assisted neurosurgical navigation system in the early 1990s, the susceptibility to ferromagnetic interference and the consecutively lower accuracy close to metal objects initially restricted its widespread use.26 Advances in technology, however, have overcome these initial limitations,27 and the latest generation of systems can detect the degree of metal interference and disable localization if expected accuracy falls beyond a critical threshold. The error of current EM systems has been reported to be in the submillimetric and thus clinically safely applicable range.24,28
Because of the small tracking volume of about 50-cm diameter and the ferromagnetic susceptibility, experience is needed to achieve a continuously stable EM navigation. The possibility of tracking without the line-of-sight issue, however, is the biggest advantage of EM systems over optic navigation. Furthermore, the small form factor that the EM coils can be built in (≤ 1-mm diameter and ≤ 10-mm length) and the possibility of subsurface tracking provide the possibility to continuously navigate the tip of small flexible instruments such as suctions, catheters, or endoscopes.23
STANDARD VS A PROPOSED ADVANCED NAVIGATION SETUP
More than 25 years after its introduction by Roberts et al,22 cranial navigation today is still most commonly performed by registration with fiducial markers, tracking with optic technology, and intermittent pointer-based intraoperative application in routine clinical settings. If navigation support is desired during a microsurgical procedure, the neurosurgeon interrupts the dissection and exchanges the current instrument for the pointer device. This navigation instrument is then introduced into the surgical site to the point of interest, and the localization of its tip is correlated with radiological imaging on the navigation screen. Care is taken to obtain an unobstructed line of sight between the pointer and the camera bar. This may require repositioning of the operating microscope or even removal of an endoscope from the operating field. Surgical instruments may be adapted for optic navigation but then are usually bulky and out of balance and therefore not widely used in intracranial microsurgery.
An advanced cranial navigation technique will seamlessly translate into the operating workflow and provide optimal accuracy. To achieve this goal, we have evaluated a setup that (1) omits the need for preoperative patient preparation and rescanning using a surface-merging registration algorithm, (2) uses EM tracking technology to avoid line-of-sight issues, (3) uses continuous tip-tracked navigation of standard neurosurgical instruments (such as suction, endoscope or biopsy needle), and (4) can be applied in an intraoperative MRI (iMRI) environment.
The aims of this study are to propose a reliable and ergonomic method of cranial neuronavigation and to provide accuracy tests, operating room setup workflow, and clinical examples.
PATIENTS AND METHODS
Since the latest generation of neurosurgical navigation systems (S7 StealthStation AxiEM; Medtronic, Louisville, Colorado) became available to the Division of Neurosurgery at the University of Calgary (Calgary, Ontario, Canada) and the Department of Neurosurgery at the Medical University of Vienna (Vienna, Austria) in September 2011, the proposed advanced navigation technique has been used clinically in 136 patients. For assessment of accuracy, comparison with current standard navigation, and evaluation of optimal operating room setup, a phantom experiment was performed at the 3.0-T iMRI facility of the University of Calgary.
Accuracy Test of Continuous EM Instrument Navigation
A phantom head made from high-density polyurethane foam (Pacific Research Laboratories Inc, Vashon, Washington) was prepared with 7 self-adhesive radiopaque fiducial surface markers (IZI Medical Products, Owing Mills, Maryland) for point-to-point registration. A cross-shaped navigation target of 2 acrylic plastic bars containing 22 surface drill holes (2-mm diameter, 2-mm depth, 1 central hole, and every 1-cm distance) was affixed in the center of the phantom head (Figure 1A). The phantom head was fixed in a metal-free skull clamp (Doro; Pro Med Instruments, Freiburg, Germany).
We used the pointer devices from the manufacturer of the navigation system for registration and targeting. For optic navigation, a pointer device with 5 infrared light-reflecting spheres on its handle was available. For EM navigation, a pen-shaped rigid pointer or a flexible wire (length, 23.7 cm) with 2 coils at its tip was used. Originally designed for shunt catheter placement,29 this so-called stylet can alternatively be inserted in hollow instruments such as suction devices, endoscopes, and biopsy needles. For testing the accuracy within a metal instrument, we inserted the EM stylet into a standard single-use metal suction (3-mm diameter, 150-mm length; Mediplast, Malmö, Sweden; Figure 1B).
A CT scan of the phantom head was acquired (General Electric Discovery CT750 HD 64-slice CT system; nontilted axial scan, 256 slices, 0.625-mm slice thickness, 512 512) and imported into the navigation system. The 7 fiducial marker positions were stored for registration.
For calculation of target accuracy, we used the trajectory measurement capability of the Cranial 2.2 software, which was adapted by the manufacturer to provide values in the submillimetric range. First, separate target points were assigned to the 22 drill holes of the acrylic bars using the CT scans. Then, the navigation probe was inserted into the drill holes, and the distance between the actual tip position and the predefined target point was measured by the system and recorded as the target error. The experiment was performed in an operating suite equipped with a ceiling-mounted 3-T MR scanner (IMRIS, Winnipeg, Manitoba, Canada). See Figure 1A for experiment setup.
Initially, the accuracy of a standard navigation setup (optic tracking, fiducial marker registration, pointer device) was calculated. Then, the advanced navigation setup was introduced stepwise to assess potential sources of inaccuracy. First, EM replaced optic tracking; then, surface merge replaced fiducial markers; and next, the tracking device was exchanged from pointer to stylet. To complete the advanced setup, we introduced the stylet into the suction tube. Then, the MR magnet was brought closer to the experiment setup (between the 5- and 50-G lines), and the accuracy tests were repeated. To simulate an iMRI scan, the EM system was removed/reattached and the accuracy checked (Table 2).
For statistical analyses, SPSS version 18.0 software (SPSS Inc, Chicago, Illinois) was used. From 3 passes of each navigation setup, the RMSE was calculated. After registration, the error was calculated by the system only for point-to-point, not for surface-based, registration. We used the Student t test to compare the target RMSE of the different steps of the advanced navigation with the standard setup. Values are presented as RMSE and range. A value of P < .05 was considered significant.
The advanced navigation was used in 136 routine cranial neurosurgical cases (Table 3) with the following setup.
The EM patient reference tracker was attached to the patient’s head via the skull clamp or directly to the skin, depending on the type of procedure. In cases performed within the iMRI suite, it was fixed to a nonferromagnetic skull clamp via a custom-made repositioning device (Figure 1A). This tool allowed realignment of the reference tracker after MRI, which requires temporary removal of the entire EM equipment. The EM field emitter was typically positioned horizontally on any side of the patient between the shoulder and skull clamp at about a 25-cm distance to reference tracker and operating field.
Multimodality Retrospective Image Application
Because registration was performed without skin fiducials, retrospective scans were routinely used. In particular, CT scanner workstations were adapted to reconstruct nontilted, 1-mm-slice-thickness image stacks from spiral acquisitions as part of their standard reconstruction protocol. This enabled us to use both emergency and routine CT examinations for surgical navigation. In addition, the T1-weighted sequence of the routine MRI protocol was adapted to whole-head 1-mm slice thickness to be used for navigation if necessary.
Patient-to-image registration was performed with the surface-based method provided by the system. Three specified points and 350 arbitrary surface points widely distributed over the patient’s head were collected with a surface probe. Because the system does not calculate registration error, anatomic landmark checks were performed routinely at 7 points (nasion and lateral canthus, philtrum/nose angle, groove medial to tragus10).
Continuous Instrument Navigation
For intraoperative application, the EM stylet described above was inserted into hollow instruments such as suction tubes, endoscopes, or biopsy needles for continuous tip-tracked instrument navigation. The wire of the EM stylet was affixed to the rubber suction tube or light/camera cable assembly of the endoscope to prevent it from crowding the operating field. Anatomic landmark checks were performed repeatedly during surgery to detect potential target error and consequently abandon navigational guidance.
Seven navigation setups were evaluated for accuracy (Table 2 and Figure 2). The standard navigation setup (optic tracking, fiducial marker registration, pointer-based navigation; setup 1) revealed an error for registration of 0.2 mm (0.2-0.3 mm) and of 0.7 mm (0.4-1.0 mm) at the target points. After the change to EM navigation, a submillimetric increase in error was observed for registration (RMSE, 0.4 mm; range, 0.2-0.5 mm) but not for targeting. During the stepwise transition to the complete advanced setup (EM tracking, surface-based registration, navigation of stylet in a metal suction tube; setup 5), no significant changes in accuracy were observed at the target points (RMSE, 0.7 mm; range, 0.3-1.2 mm).
When the experiment was performed closer to the iMRI magnet (within the 5-G and outside the 50-G lines; setup 6), we observed a significant decrease in accuracy (RMSE, 0.9 mm; range, 0.7-1.3 mm). Accuracy decreased even more when the patient reference tracker was temporarily removed during acquisition of an iMRI and subsequently repositioned in its holder (setup 7).
In summary, no significant difference in target accuracy was noted between standard navigation and the proposed advanced navigation setup when performed outside the iMRI 5-G line.
Continuous EM instrument navigation was feasible and accurate in all but 6 of 136 cases, which were performed during the initial month after the installation. After an initial learning curve, no difference in setup time was found between the standard and advanced navigation setups.
In 3 of the 6 cases, the patient reference tracker was not traceable because of its proximity to the metallic skull clamp (2 prone positions, 1 sitting position). This error has not occurred again since we began to use either a custom-made acrylic cube that generates 4-cm distance between the reference tracker and metal skull clamp or a nonferromagnetic skull clamp. Furthermore, in positions other than supine, the EM field generator may be removed from its holding device to facilitate the registration procedure.
Nonrigid Fixation of Patient Reference Tracker
In 2 of the 6 cases (1 endoscopic transsphenoidal surgery, 1 tumor cyst catheter placement), the skin-attached patient reference tracker performed some movement after draping (1 case by Galea movement owing to the weight of the drape, 1 case by dissolution of the glue owing to the skin preparation with alcohol). Attaching the patient reference tracker to the skull clamp or firmly fixing it to the forehead with bandage has prevented reoccurrence of these errors.
Patient Movement in the Skull Clamp
In the last of the 6 cases of navigation error, an awake surgery was performed. During removal of the laryngeal mask, some head movement occurred relative to the skull clamp to which the patient reference tracker was affixed and caused considerable navigation error. Since then, we have used the bone screw-attached reference tracker for awake procedures.
Besides catheter placement (n = 9), which is the approved indication of EM stylet navigation and is described elsewhere,30,31 continuous EM instrument navigation was used in the following procedures.
Intracranial Microsurgical Tumor Resection
Continuous suction navigation during tumor resection provided the performing surgeon with constantly updated information about the present working position (n = 71; Figure 3A). During fluorescence-guided resection of a malignant glioma, the surgeon is tempted to remove as much fluorescing tissue as possible. Continuous EM instrument navigation with diffusion tensor imaging tractography constantly visualized the proximity of the resection to eloquent fiber tracts, guided subcortical stimulation, and prevented the surgeon from inadvertent removal of fluorescing but functioning white matter areas (Figure 3A). The issue of brain shift was anticipated by accuracy checks on anatomic landmarks such as vessel bifurcations and sulci.
In 8 cases of microsurgical tumor resection, the EM stylet was mounted to the suction tool of the neuroArm neurosurgical robot (IMRIS). The neurosurgeon controlling the robot was able to observe the current robot working position onscreen at the workstation outside the operating room (Figure 3B).
Endoscopic Transsphenoidal Surgery
The advantages of navigation over fluoroscopy in transsphenoidal surgery have been defined,32 although the line-of-sight issue when using optic navigation concurrently with an endoscope in a single nostril prevented continuous navigation in most cases. EM suction navigation has completely overcome this line-of-sight issue. Furthermore, a firmly affixed adhesive patient tracker on the forehead obviated the need for rigid fixation of the patient’s head in a skull clamp, which provided head mobility during surgery and added to patient comfort postoperatively (n = 46; Figure 3C).
In cases of endoscopic third ventriculostomy, navigated insertion of the endoscope along a trajectory through the foramen of Monroe into the third ventricle is crucial to avoid procedural morbidity by lesioning the thalamus or fornix.33 We used the neuronavigation system to plan the optimal point of entry at the skull as the back-projection of the line from the foramen of Monroe (entry point) and the floor of the third ventricle between the mammillary bodies and basilar bifurcation (target point). By inserting the EM stylet into the working channel of the endoscope, the navigation system provided the performing surgeon with a continuous update of the position of the endoscope tip in relation to the trajectory line. This was particularly helpful when advancing the scope through the frontal white matter before piercing the ependyma of the frontal horn. Once the endoscope reached its optimal position inside the third ventricle, the stylet was advanced and its tracked tip used to perform the perforation at the target point under continuous image guidance (n = 6; Figure 3D). The opening was then enlarged with a balloon catheter.
An identical setup was used for the management of arachnoid and ventricular cysts. The navigation system provided the possibility for preoperative planning of a trajectory and continuous intraoperative orientation, and the EM stylet was used to pierce cyst membranes under navigational guidance (Figure 3E).
In 4 cases, we successfully used the EM navigation to guide the stereotactic biopsy needle by inserting the stylet into a standard side-cutting needle.
According to our data, the proposed advanced navigation technique of continuous EM instrument navigation was superior in terms of neurosurgical workflow ergonomics compared with standard navigation with fiducial markers, optic tracking, and pointer-based application because of the following aspects: It obviates the need for preoperative fiducial marker application and rescanning, eliminates the line-of-sight issue of optic tracking, and enables continuous on-screen update of standard instruments such as suction tubes, endoscopes, or biopsy needles. The phantom experiment further supported this concept, confirming that continuous EM instrument navigation provides accuracy equal to that of standard navigation even when applied to surgeries within an iMRI environment within the safety zone outside the 5-G line.
Integration Into Surgical Workflow
The ergonomic advantage of the presented setup lies in the seamless integration into the surgical workflow. Deep-seated lesions may be approached with continuous update of the current instrument position on imaging. While the surgeon operates with the accustomed suction fitted with the EM stylet, the tip of the suction continually updates on the navigation screen, always providing information about the distance to the tumor border and surrounding structures.
In skull base surgery, even minor distortions of the suction tube shaft owing to bone dissection maneuvers do not render navigation inaccurate because our EM setup uses tip tracking. With such tip-tracked continuous instrument navigation, the surgeon can continue the dissection as needed and can check the position of the instrument on the navigation screen at any time. In contrast, in standard optic pointer-based navigation, the surgeon has to interrupt dissection, exchange the current instrument with the navigation pointer, and check for free line-of-sight. In automotive navigation, this would correspond to a driver stopping the car, activating a navigation device, and adjusting the satellite reception before the current position could be checked.
The EM stylet can be both inserted into the suction tube and introduced into the working channel of an endoscope. In ventriculostomy cases, the endoscope can then be advanced under EM guidance through the interventricular foramen with the same precision as a stereotactic biopsy needle. Once the endoscope is in the appropriate position inside the third ventricle, the EM stylet can be advanced further to puncture the target point under direct endoscopic view and EM guidance.
Previous studies on the different methods of registration using optic tracking have shown that besides bone screws (error, 0.23 ± 0.03 mm under laboratory conditions8), which are not applicable in the routine clinical setting, skin fiducial marker registration provides the highest accuracy (error, 1.1-4.0 mm).9,34,35 Registration relying solely on anatomic landmarks had the lowest accuracy (3.2-3.9 mm),9,10 and registration based on surface points was found to provide intermediate accuracy (3.3 ± 1.65 mm).9 Our phantom accuracy experiment revealed an equally low calculated error for registration (mean error, 0.2-0.4 mm) with optic and EM navigation.8 Our submillimetric higher mean target error of 0.7 mm corresponds well to the previous laboratory experiments,8,24 given that we used fiducial marker or surface-based registration, not bone screws. We did not find any significant difference in target error between fiducial marker and surface merge registration.
Previous studies have reported higher accuracy with the use of CT scans for patient registration compared with MRIs because of small inhomogeneities of the magnetic field.10,36 In cases when high accuracy was needed such as frameless biopsies of small targets, we always used a fusion of CT scan for registration and MR for target selection (Figure 3E).
Few studies comparing optic and EM tracking exist. In the experiment of Kral et al,24 optical tracking was significantly more accurate than EM tracking (median target error, 0.12 vs 0.37 mm, respectively; P < .001). However, they used fiducial marker registration and bone-affixed screws as targets. In contrast, we did not find a significant difference between optical and EM tracking (mean target error, 0.7 vs 0.6 mm, respectively). In our experience, evaluation of accuracy in the submillimetric range is limited by the display resolution when target points are defined manually. It is of note that the highest accuracy (error, ≤ 0.5 mm) was always found in the center of the phantom, whereas the highest error (up to 1.2 mm) was encountered in the target points at the periphery of the EM field. Therefore, we recommend positioning the EM emitter approximately 25 cm distant and pointing to the center of the surgical target for highest accuracy.
Although navigation of instruments with the EM stylet inside metal tubes has been reported,37 we are unaware of literature reporting the inaccuracy of this setup. Our results show equal accuracy for the standard navigation and our advanced navigation setup. Furthermore, this is the first report on accuracy tests of EM navigation in an iMRI environment. Outside the 5-G line, no significant difference between optic and EM navigation was observed. As expected, the higher magnetic field (just inside the 5-G line) led to decreased accuracy of the EM navigation.
Setup and Learning Curve
The introduction of EM navigation has a learning curve. This is reflected by the erroneous 6 cases in our series, all of which occurred within the first months of the experiment. Within the scope of this project, we have acquired knowledge about the optimal setup of EM navigation. First, no metal parts should reside between the emitter and patient tracker. Although inaccuracy from ferromagnetic interference is prevented by the navigation system, it stops the system from working if too extensive. Metal skull clamps, wires in the surgical drape, and instruments such as retractors or specula for transnasal surgery cause considerable interference with the EM field, even if they are MRI compatible and made of titanium. Second, the EM field emitter does not need to be fixed to the patient’s head but can be manually readjusted during registration or during surgery in cases of bad communication with the system. Third, the patient reference tracker needs to be firmly fixed to the patient’s head throughout the procedure. If no rigid head fixation with a skull clamp is desired, this can be achieved by a skull-mounted tracker via 2 bone screws. Because of its considerable invasiveness, this setup should be reserved for selected cases in which a skull clamp is not applicable and high accuracy is needed such as in awake surgeries. Alternatively, a skin-adhesive tablet-shaped patient tracker is available. Fixing this device to the skin, however, requires some expertise because the skin over certain areas of the skull may move considerably owing to the weight of the surgical drape or even the surgeon’s hands and instruments. If the patient’s head is fixed in a skull clamp, the adhesive patient tracker can be attached to the clamp either with a distance of approximately 4 cm (in case of a metal skull clamp) or directly (in case of nonferromagnetic clamp). We routinely use the latter configuration because it provides maximum accuracy. Finally, although our study shows that EM navigation can be used safely and accurately in the iMRI environment, execution of an intraoperative scan requires removal of all parts of the EM navigation setup. Although the position can be marked or a holding device can be left in place (Figure 1A), it is of note that reattachment of the patient reference tracker is prone to considerable inaccuracy.
Dedicated EM Instruments
Because EM navigation is relatively new to the field of neurosurgery, current equipment can be improved, and the development of dedicated EM instruments is necessary. A minimally invasive patient tracker version that can be fixed to the patient’s head by a single screw via a stab incision should be considered for the future. Furthermore, a method of precisely reattaching the patient tracker after iMRI imaging needs to be developed.
Inserting the EM stylet into the suction requires a certain diameter (≥ 3 mm) that prevents the tube from being blocked repeatedly by aspirated material. Furthermore, cleaning of the navigated suction device should be performed by injection of saline into the suction tip because multiple removals of the stylet can break the thin wires to the EM coils on the stylet tip. Therefore, we advocate the design of dedicated EM instruments for neurosurgery. Although such instruments exist for ear, nose, and throat (otolaryngology) and maxillofacial surgery, they are not tip tracked; rather, the coil resides on the handlebar, which distorts small instruments during dissection. The optimal microneurosurgical suctions would be of variable diameter and include the EM coils around the tip wall of the suction tube.
Economic Considerations and Indications for Application
A drawback of the proposed setup is the considerable higher cost per case of the EM setup compared with optic navigation. Although we tried to illustrate how the advantages of tip-tracked continuous instrument navigation may outweigh the additional financial burden, strict indications should be instituted for its application. If navigation is used only to outline the target circumference for planning the skin incision and craniotomy (in cases of superficial pathologies such as convexity meningiomas), optic navigation seems more justified from an economic view. Furthermore, the EM navigation technique has recently been proposed for frameless stereotactic biopsy because it may obviate the need for rigid head fixation and may be performed in local anesthesia.37 Although this seems warranted in selected cases of contraindication/high risk to general anesthesia, optic tracking should be considered in standard biopsy cases performed under general anesthesia because of its lower cost. As the number of EM applications rises, cost per case should drop accordingly in the future.
Continuous instrument navigation is the prerequisite for seamless integration of navigation systems into the neurosurgical operating workflow. Our data confirm that the application of preoperative imaging, surface-merge registration, and continuous EM tip-tracked instrument navigation provides such integration without a significant reduction in accuracy compared with standard optic navigation with skin fiducials. Furthermore, the proposed advanced navigation setup was tested with equally high accuracy in the safety zone of the iMR environment outside the 5-G line. However, technical refinements of navigated instruments and of patient reference trackers are required.
This work has been supported in part by an educational grant of Medtronic Navigation. Dr Wolfsberger is currently an educational consultant and a technological advisory board member of Medtronic. The other authors have no personal financial or institutional interest in any of the drugs, materials, or devices described in this article.
We wish to thank Ingrid Dobsak for preparation of the figures and Claire Lacey for manuscript assistance.
1. Perneczky A, Reisch R, Tschabitscher M. Keyhole Approaches in Neurosurgery: Concept and Surgical Technique. Wien, Austria: Springer-Verlag; 2008.
2. Apuzzo ML, Chen JC. Stereotaxy, navigation and the temporal concatenation. Stereotact Funct Neurosurg. 1999;72(2-4):82–88.
3. Grunert P, Darabi K, Espinosa J, Filippi R. Computer-aided navigation in neurosurgery. Neurosurg Rev. 2003;26(2):73–99; discussion 100-101.
4. Fitzpatrick JM. The role of registration in accurate surgical guidance. Proc Inst Mech Eng H. 2010;224(5):607–622.
5. Steinmeier R, Rachinger J, Kaus M, Ganslandt O, Huk W, Fahlbusch R. Factors influencing the application accuracy of neuronavigation systems. Stereotact Funct Neurosurg. 2000;75(4):188–202.
6. Fitzpatrick JM. Fiducial registration error and target registration error are uncorrelated. Proc SPIE. 2009;7261:726102.
7. Maurer CR Jr, Fitzpatrick JM, Wang MY, Galloway RL Jr, Maciunas RJ, Allen GS. Registration of head volume images using implantable fiducial markers. IEEE Trans Med Imaging. 1997;16(4):447–462.
8. Brinker T, Arango G, Kaminsky J, et al.. An experimental approach to image guided skull base surgery employing a microscope-based neuronavigation system. Acta Neurochir (Wien). 1998;140(9):883–889.
9. Pfisterer WK, Papadopoulos S, Drumm DA, Smith K, Preul MC. Fiducial versus nonfiducial neuronavigation registration assessment and considerations of accuracy. Neurosurgery. 2008;62(3 suppl 1):201–207.
10. Wolfsberger S, Rössler K, Regatschnig R, Ungersböck K. Anatomical landmarks for image registration in frameless stereotactic neuronavigation. Neurosurg Rev. 2002;25(1-2):68–72.
11. Ende G, Treuer H, Boesecke R. Optimization and evaluation of landmark-based image correlation. Phys Med Biol. 1992;37(1):261–271.
12. Pelizzari CA, Chen GTY. Registration of multiple diagnostic image scans using surface fitting. In: The use of computers in radiation therapy. Elsevier, North Holland. 1987;437–440.
13. Besl P, Mckay N. A method for registration of 3-D shapes. IEEE Trans Pattern Anal Mach Intell. 1992;14(2):239–256.
14. Marmulla R, Mühling J, Wirtz CR, Hassfeld S. High-resolution laser surface scanning for patient registration in cranial computer-assisted surgery. Minim Invas Neurosur. 2004;47(2):72–78.
15. Schicho K, Figl M, Seemann R, et al.. Comparison of laser surface scanning and fiducial marker-based registration in frameless stereotaxy: technical note. J Neurosurg. 2007;106(4):704–709.
16. Raabe A, Krishnan R, Wolff R, Hermann E, Zimmermann M, Seifert V. Laser surface scanning for patient registration in intracranial image-guided surgery. Neurosurgery. 2002;50(4):797–801.
17. Shamir RR, Freiman M, Joskowicz L, Spektor S, Shoshan Y. Surface-based facial scan registration in neuronavigation procedures: a clinical study. J Neurosurg. 2009;111(6):1201–1206.
18. Watanabe E, Watanabe T, Manaka S, Mayanagi Y, Takakura K. Three-dimensional digitizer (neuronavigator): new equipment for computed tomography-guided stereotaxic surgery. Surg Neurol. 1987;27(6):543–547.
19. Watanabe E, Mayanagi Y, Kosugi Y, Manaka S, Takakura K. Open surgery assisted by the neuronavigator, a stereotactic, articulated, sensitive arm. Neurosurgery. 1991;28(6):792–799; discussion 799-800.
20. Adams L, Krybus W, Meyerebrecht D, et al.. Computer-assisted surgery. IEEE Comput Graph. 1990;10(3):43–51.
21. Maciunas RJ, Galloway RL Jr, Fitzpatrick JM, Mandava VR, Edwards CA, Allen GS. A universal system for interactive image-directed neurosurgery. Stereotact Funct Neurosurg. 1992;58(1-4):108–113.
22. Roberts DW, Strohbehn JW, Hatch JF, Murray W, Kettenberger H. A frameless stereotaxic integration of computerized tomographic imaging and the operating microscope. J Neurosurg. 1986;65(4):545–549.
23. Glossop ND. Advantages of optical compared with electromagnetic tracking. J Bone Joint Surg Am. 2009;91(suppl 1):23–28.
24. Kral F, Puschban EJ, Riechelmann H, Pedross F, Freysinger W. Optical and electromagnetic tracking for navigated surgery of the sinuses and frontal skull base. Rhinology. 2011;49(3):364–368.
25. Kato A, Yoshimine T, Hayakawa T, et al.. A frameless, armless navigational system for computer-assisted neurosurgery: technical note. J Neurosurg. 1991;74(5):845–849.
26. Birkfellner W, Watzinger F, Wanschitz F, et al.. Systematic distortions in magnetic position digitizers. Med Phys. 1998;25(11):2242–2248.
27. Schicho K, Figl M, Donat M, et al.. Stability of miniature electromagnetic tracking systems. Phys Med Biol. 2005;50(9):2089–2098.
28. Hummel J, Figl M, Birkfellner W, et al.. Evaluation of a new electromagnetic tracking system using a standardized assessment protocol. Phys Med Biol. 2006;51(10):N205–N210.
29. Hayhurst C, Byrne P, Eldridge PR, Mallucci CL. Application of electromagnetic technology to neuronavigation: a revolution in image-guided neurosurgery. J Neurosurg. 2009;111(6):1179–1184.
30. Kandasamy J, Hayhurst C, Clark S, et al.. Electromagnetic stereotactic ventriculoperitoneal CSF shunting for idiopathic intracranial hypertension: a successful step forward? World Neurosurg. 2011;75(1):155–160; discussion 32-33.
31. Hayhurst C, Beems T, Jenkinson MD, et al.. Effect of electromagnetic-navigated shunt placement on failure rates: a prospective multicenter study. J Neurosurg. 2010;113(6):1273–1278.
32. McGrath BM, Maloney WJ, Wolfsberger S, et al.. Carotid artery visualization during anterior skull base surgery: a novel protocol for neuronavigation. Pituitary. 2010;13(3):215–222.
33. Rohde V, Behm T, Ludwig H, Wachter D. The role of neuronavigation in intracranial endoscopic procedures. Neurosurg Rev. 2012;35(3):351–358.
34. Wolfsberger S, Czech T, Knosp E. Pituitary adenomas: neurosurgical treatment [in German]. Wien Klin Wochenschr. 2003;115(suppl 2):28–32.
35. Paraskevopoulos D, Unterberg A, Metzner R, Dreyhaupt J, Eggers G, Wirtz CR. Comparative study of application accuracy of two frameless neuronavigation systems: experimental error assessment quantifying registration methods and clinically influencing factors. Neurosurg Rev. 2010;34(2):217–228.
36. Maciunas RJ, Fitzpatrick JM, Gadamsetty S, Maurer CR Jr. A universal method for geometric correction of magnetic resonance images for stereotactic neurosurgery. Stereotact Funct Neurosurg. 1996;66(1-3):137–140.
37. Harrisson SE, Shooman D, Grundy PL. A prospective study of the safety and efficacy of frameless, pinless electromagnetic image-guided biopsy of cerebral lesions. Neurosurgery. 2012;70(1 suppl operative):29–33; discussion 33.