Craniotomy and Clipping of Intracranial Aneurysm in A Stereoscopic Virtual Reality Environment

OBJECTIVE: The release of results of International Subarachnoid Aneurysm Trial in 2003 caused a shift in the paradigm of management of ruptured intracranial aneurysms. The cases selected for microsurgical clipping nowadays are usually those patients with aneurysms that are not suitable for embolization, and are often complex and difficult. We devised an innovative application of operative planning and training for craniotomy and microsurgical clipping of intracranial aneurysms in a stereoscopic virtual reality environment.

Cited Here...: Patient-specific Digital Imaging and Communications in Medicine data from computed tomographic angiography of the intracranial circulation and cranium were transferred to the workstation (Dextroscope; Volume Interactions Pte. Ltd., Singapore, Singapore). An aneurysm clip database was loaded into the patient data set. Three-dimensional volume rendering was followed by data coregistration and fusion.

Cited Here...: Virtual head positioning and craniotomy were carried out to simulate the microscopic visualization. Clip selection could be carried out with reference to the angle of application. This allows one to see the exposure and degree of obliteration of an aneurysm with the various angles of approach.

Cited Here...: The virtual craniotomy and microsurgical clipping application simulated the operative environment. Its role in neurosurgical training is encouraging and should be further developed.

The microsurgical clipping of intracranial aneurysms is a crucial part of vascular neurosurgery. The release of the results of the International Subarachnoid Aneurysm Trial in 2003 caused a shift in the paradigm of the management of ruptured intracranial aneurysms (^17,18). Although this study suffered from a number of weaknesses in methodology and criticism in applicability, the management of ruptured and unruptured intracranial aneurysm has swung from microsurgical clipping to endovascular embolization as the preferred mode of treatment in many neurosurgical centers (^6,15). The aneurysm patients who are referred for microsurgical clipping are usually those with aneurysms that are not suitable for embolization, which are sometimes complex and difficult. Neurosurgical training for microsurgical clipping becomes a great challenge.

Virtual reality simulation has long been used for training in military and civilian arenas, such as pilot training. The advantage of training in a no-risk virtual reality environment is appreciated by the surgical community. Recently, simulators have been developed for laparoscopic surgery, endoscopy, and interventional radiology. A similar application in neurosurgery (Dextroscope; Volume Interactions Pte. Ltd., Singapore, Singapore) is also available. This system uses patient-specific, coregistered, fused radiology data sets that may be viewed stereoscopically and can be manipulated in a virtual reality environment. It allows some degree of rehearsal of bone drilling and brain retraction using the erase function. The use of the Dextroscope in intracranial tumor cases was demonstrated previously (¹⁰). The application of the Dextroscope to operative planning and training for craniotomy and the microsurgical clipping of intracranial aneurysm has not been tested and described. We have developed a means to provide instruments and implants for preoperative planning and rehearsal of microsurgical clipping of intracranial aneurysms using the Dextroscope-based virtual reality environment.

MATERIALS AND METHODS

The details of the Dextroscope have been described previously (¹⁰). Briefly, a computer-generated stereoscopic image is displayed on a monitor (1024 × 768 pixels) and is reflected into the user's eyes via a mirror. The user works from a seated position, with his or her forearms positioned on comfortable armrests. Wearing liquid crystal display shutter glasses (Crystal Eyes; Sterographics Inc., San Rafael, CA), which are synchronized with the time-split display, the user looks into the mirror and sees a stereoscopic image that seems to float behind the mirror. The positions of the virtual holder and stylus in the user's hands are calibrated so that the user perceives them in the same spatial position and orientation.

Patient-specific radiology data sets were processed as described. A Standard 16-row multislice computed tomographic (CT) scanner (GE LightSpeed 16; GE Healthcare, Waukesha, WI) was used to obtain three-dimensional CT angiography data sets. Contrast solution (Omnipaque 300 [GE Healthcare], 100ml, 3.5-4m l/s) was given with a delay of 20 to 25 seconds. Data was acquired with a thickness of 0.625 mm from the foramen magnum to the vertex. Digital Imaging and Communications in Medicine data using patient-specific CT scans of the cranial bone and CT angiography of intracranial circulation were loaded into the Dextroscope workstation. Three-dimensional volume rendering was followed by data coregistration and fusion. A three-dimensional color-adjustment table allowed individual color and transparency adjustments for the cranial bone and cerebral arteries. Structures, such as aneurysms, could be selected, shaped, and colored. The process was straightforward because all data were CT-based.

Aneurysm clips and clip applicators were scanned using a standard 16-row multislice CT (GE LightSpeed 16). Using the Dextroscope, the data were processed by cropping out the empty space and useless information. Further coloring could then be performed. The data sets were saved as the aneurysm clip database. The aneurysm clip database could then be loaded into an individual patient data set.

RESULTS

The patient-specific data set was transferred, colored, and highlighted. The user could then visualize the anatomy with various available functions, including zoom, rotate, move, crop, and cut plane. The measurement function could be activated to measure the dimension of aneurysm.

Using the crop function for the cranial bone, the best angle and position could be selected with respect to clip placement at this juncture (Fig. 1A). The selected angles of positioning could be visualized from the left-sided logo, and that position would be the ideal position for the microscope.

Figure 1

The user then returned to the complete cranial view using the uncrop feature and positioned the head in accordance with the selected angles of positioning. Craniotomy can then be performed using the drill function. In the case of a right middle cerebral artery aneurysm, a pterional craniotomy was performed (Fig. 1B). The user could then provide feedback regarding the adequacy of different extension of the craniotomy and the effect of drilling of the sphenoidal ridge and temporal bone with respect to the visualization of the aneurysm, internal carotid artery, and middle cerebral artery.

The method used to bring in the clip and clip applicator is described here. In the registration session, the user could use freehand registration for the CT angiography and the suitable clips. One can then move the clip to the neck of aneurysm as appropriate (Fig. 1C). The same principle of freehand registration could be performed to the clip and clip applicator beforehand. In that case, the subsequent visualization of the aneurysm clip applicator could tell the user whether or not the clip applicator would be obstructing the working microscopic view in reality (Fig. 1D). By repeating the procedures in different angles and degrees of zoom, the user could gain a better understanding of the best approach for microsurgical clipping for the patient.

As in the above case of a middle cerebral artery aneurysm, we performed similar work in other data sets, including patients with internal carotid artery aneurysms (Fig. 2), anterior communicating artery aneurysms (Fig. 3), and vertebral artery aneurysms, amounting to a total of 13 cases. Difficult cases, such as a giant internal carotid artery aneurysm, were useful to obtain a better appreciation of the anatomy and neurosurgical approach.

Figure 2

Figure 3

DISCUSSION

Technology has improved our methods of understanding neurosurgical anatomy and approaches. With the advance of three-dimensional angiography technology, one can easily manipulate three-dimensional intracranial vasculature in the computer console, whether from CT angiography, magnetic resonance angiography, or catheter angiography (^2,16,20,23). Advances in neuronavigation technology allow neurosurgeons to look at anatomic structures and lesions in a different plane, which allows better decision making for neurosurgical approaches (^7,26). Neurosurgeons can now manipulate a two-dimensional projection of a three-dimensional cerebral vasculature with a different degree of rotation and magnification. This has certainly improved the neurosurgeon's preoperative anatomic understanding of the cerebral vasculature and pathology. The projected views, however, are still significantly inferior, in terms of visual quality, to the intraoperative binocular microscopic views. Traditional training methods for microsurgical clipping involve training of microsurgical techniques in animals, case observations, and a step-by-step increase in participation during the operation. This tradition of training has raised concern among professionals and the public. The problem is further complicated by a decrease in case load, which reduces the frequency of opportunities for neurosurgical trainees to observe and participate in the operation. Worse still is that the aneurysms now faced by the neurosurgeons would be the aneurysms that are not suitable for coiling, which may be large, wide-necked, and partially thrombosed aneurysms associated with intracranial stenosis or with important vessels coming out from the aneurysm complex. Virtual reality simulation has long been used for training in military and civilian arenas, such as air pilot training. The advantage of training in a no-risk virtual reality environment is appreciated by the surgical community. Recently, simulators have been developed for laparoscopic surgery, endoscopy, and interventional radiology (^{3-5,9,19,21,22}). A similar application, the Dextroscope, is also available in neurosurgery. Virtual reality simulation seems to have come at the right time to provide a good solution for the appropriate training of the microsurgical clipping of aneurysms.

The Dextroscope allows real-time manipulation of the three-dimensional data set and provides a suite of registration, segmentation, and planning tools, including tools for simulating bone drilling. The three-dimensional views obtained from the stereoscopic virtual reality environment are in keeping with what one attained with the binocular microscopic views, as compared with the two-dimensional projections of the three-dimensional CT angiography. We further enhance its applicability by incorporating aneurysm clips and aneurysm clip applicators to the data set, in which movement in relationship to the three-dimensional CT angiography data (and, thus, the aneurysm) is feasible. It allows the option of preoperative rehearsal and training for microsurgical clipping.

The use of virtual reality simulation for the training of microsurgical clipping allows for the development of a pretrained novice so that basic psychomotor and visual-spatial skills, with reference to positioning, craniotomy, and clip application, have been developed and become reasonably automatic before proceeding to the operating room. In conjunction with cadaveric dissection training for microsurgical skills, this allows for more efficient and safer use of operating room training time, such that intraoperative training can be focused on avoidance of errors and other technical perils. However, when one considers the implementation of virtual reality simulation into the training program, it is important to realize that virtual reality simulation is most effective when incorporated into a well-designed training program of vascular neurosurgery with close supervision and participation of vascular neurosurgeons. A high fidelity laparoscopic training model has been shown to improve resident performance of basic skills in the animate operating room environment and provides evidence that virtual reality simulation training may lead to improved operating room performance (¹). The next stage would be to develop a validated assessment tool for operative performance of microsurgical clipping and to assess the results of microsurgical clipping training with or without the availability of the stereoscopic virtual reality environment. Effectiveness, in terms of operative time and clinical outcome, would be another important aspect to be explored in the next stage of study.

The other spectrum of utilization would be for the vascular neurosurgeons to understand complex vascular cases. This virtual reality system allows the appreciation of the anatomy from different angles. It can highlight different areas of interest, such as the aneurysm, main trunk, important branches, and the relationship to bony structures. It allows better preoperative planning, better decisions regarding operative approaches, and a preoperative rehearsal. It may be used for the training and decision to use the keyhole approach. The system can also be used in the clinic and ward settings to facilitate the acquisition of informed consent for the operative procedures, allowing patients a view to help relieve anxiety and build confidence.

The same principle could be applied to other vascular cases, including cerebral arteriovenous malformations. It provides better appreciation of the arterial and venous anatomy and better pre-operative planning. However, the current hardware and software cannot account for the many intraoperative variations that can change management, including brain swelling, brain retraction, movement of structures on manipulation, and calcification of the aneurysm neck, that could alter clip application and aneurysm rupture. Future development would involve the simulation of brain retraction, clip opening and closure with the associated aneurysm deformation, haptics, simulation of intraoperative aneurysmal rupture, and so on (^11-14,24,25). We believe that it will ultimately become an indispensable tool for neurosurgical training and planning in every major vascular neurosurgery center (⁸).

CONCLUSION

The virtual microsurgical clipping application enhances the understanding of the surgical anatomy and simulates the operative environment encountered in microsurgical clipping. It provides an opportunity for preoperative rehearsal, microsurgical training, and assessment.

Disclosure

George K.C. Wong and W.S. Poon are honorary key opinion leaders for Volume Interactions Pte. Ltd. and provide input for future research and development. None of the authors have received any direct funding from Volume Interactions Pte. Ltd.