Computer-assisted Spinal Surgery for Deformity – A Review
The specialities of both orthopaedics and neurosurgery have been forerunners in the implementation of the principles of image-guided surgery including robotic surgery. Spinal surgery was thus well placed for the application of the principles of computer- assisted surgery. Nolte et al. at Bern University in Switzerland developed the concept of interactive navigation of surgical instruments in spinal surgery. Merloz et al. at Grenoble University in France developed the concept of anatomically based registration using only reference anatomical structures detected by intra-operative sensors.18
Simon and Foley developed this technology in North America.19,20
With developments in the area of image processing and understanding, newer concepts of image segmentation, 3D model building and registration were developed, which formed the basis of frameless computer-assisted surgery. Notable works in the field of imaging included that of Brown19
in 199221 registration followed by Van den Elsen22–26
on 2D photometric image on 3D-image registration.
Overview of the Technical Aspects of Computer-navigated Spinal Surgery It has been stated that CNSS is a man–machine interface aiming to accomplish delicate tasks in spinal surgery that require precision to avoid damage to the adjacent neurovascular structures, to achieve the surgical objective with the highest accuracy and to minimise tissue dissection in order to bring about sound healing. This is achieved by integration of anatomical imaging, visual feedback and surgical dexterity.4
An extension of this, as noted earlier, is also to attain an optimal implant position, hence optimising post-operative biomechanics and biological effects. This is particularly important in the context of spinal deformity. The technical aspects and basic concepts of navigation have been extensively reviewed. We shall therefore briefly summarise the overall principles for completeness.2,3,8,15,20,27
The underlying concepts of this approach are quite simple. The overall aim of this procedure is to show the operating surgeon the exact position of a hand-held instrument in relation to the bony anatomy, represented on an image displayed in the operating theatre. Although there is a variety of specific technologies that act to link, process and display the relevant information in each type of navigation, the underlying methods are similar.
Current CNSS can be considered as a duality of image acquisition and processing followed by intra-operative navigation. Image processing can be pre-operative or intra-operative and is a three-step process. Image acquisition is where either a pre-operative computerised tomography (CT) scan or an intra-operative image, either with fluoroscopy or ultrasound, is acquired in the computer. Referencing must then occur in which remote tracking technologies such as infrared light-based optoelectronic tracking, electromagnetic tracking or ultrasound-based tracking are used to form a relationship between intra-operative position and orientation of tools used in the process.3,9,10,12,28
Surface fiducials or a dynamic reference array (DRA) attached to the spinous process allow constant tracking of the location and orientation of the vertebra. Finally, the process of registration involves correlation (matching) of the anatomical landmarks on the subject with the image that has been processed.
EUROPEAN MUSCULOSKELETAL REVIEW
reference device to perform invasive procedures, has been around since 1873 when Dittmar developed and reported guiding devices for the placement of probes into the medulla oblongata of animals. Its clinical use was developed by the pioneering work performed by Robert H Clarke and Victor Horsley in 1906 for treating intracranial pathology.16,17
Registration can be paired point matching and/or surface matching. This leads to the development of a surgical plan determining the trajectory and linear dimensions of the pedicle.
intra-operative images are used to register the patient’s anatomy to the pre-operative stored CT scan, or an intra-operative fluoroscopy image or isocentric motorised fluoroscope provides the fully registered image.
A concern with registration is that, unless multiple DRAs are applied to each vertebra and time-consuming registration applied, any movement between vertebra will affect accuracy. For CT-based navigation, this can be reduced by trying to ensure that the pre-operative CT position mimics the position on the operating table as much as possible. However, in standard pre-operative CT-based systems, any movement during theatre and the effects of surgery itself will not be updated on the CT image. These problems are alleviated by the use of a C-arm with a reference array attached or an isocentric motorised C-arm. In these techniques,29
CT-based navigation using optoelectrical tracking allows 3D navigation; however, the major disadvantage is the amount of radiation exposure due to the requirement for pre-operative thin-sliced, high-resolution CT images. The other disadvantages cited are a significant learning curve, particularly with the registration process, and, as previously stated, the relative difference in the anatomy between the pre-operative CT scans taken in supine position and the intra-operative prone position. The latter disadvantage can particularly be a hindrance in cases of complex spinal deformity or instability. These problems have led to the development of intra-operative CT-based image guidance. However, intra-operative CT-based guidance systems are expensive and need a specially designed operating table which may limit their use to highly specialised spinal centres.30
Discussion of Navigation Strategies and their Development
Image processing and developing a surgical plan are followed by the stage of navigation in which execution of the surgical plan is carried out. Navigation can be either CT- or fluoroscopy-based as stated above.31,32
Fluoroscopy-based navigation can be either in 2D or 3D mode.
Foley (1996) developed an interactive intra-operative image-guided approach based on the concept of virtual fluoroscopy that combined the use of standard C-arm fluoroscopy with computer-aided surgical technology to improve the registration process.20
The advantage of such
a system was that it allowed the changing of registration points until acceptable accurate matching with the pre-operative image was obtained. The relative deformation between the skin surface and the underlying vertebral anatomy, observed by Brodwater (1993), was overcome as the DRA was attached to the spinous process, giving realtime information about the position of the individual vertebra. The main advantage of such a system is the relative familiarity of the fluoroscope to the operating surgeon, as it is widely used in various surgical procedures in both neurosurgery and orthopaedics. By avoiding the need to take multiple images to update the realtime position of the spine, the exposure of the operating team and patient to radiation is minimised. Virtual fluoroscopy also reduces the amount of time taken to perform the surgery as all the images are taken at the beginning of the procedure and frequent repositioning of the image intensifier is avoided. The main disadvantage of virtual fluoroscopy is that it provides 2D-image guidance and therefore the procedure is still dependent on the experience of the surgeon. Poor imaging due to various factors, such as
49
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68