Published 4/1/2011
S. Raymond Golish, MD, PhD; Paul A. Anderson, MD

Cone-beam CT enhances spinal navigation

Intraoperative C-arm fluoroscopy has evolved to allow rapid production of axial images by computed tomography (CT). Traditional C-arm units rotate in a cam shape that does not create a fixed center; with isocentric C-arm technology, the X-ray emitter rotates about a fixed center on the patient, who is positioned on a radiolucent table. As the C-arm rotates, it takes conventional images from a typical cone-shaped X-ray beam, which can be used to reconstruct CT images in any plane. Three-dimensional reconstruction is also possible. Further, cone-beam CT technology can be adopted to aid surgical navigation.

Better registration
One limitation in traditional spinal navigation was the need for manual intraoperative registration of surgical anatomy with either preoperative CT data or intraoperative standard fluoroscopy images. Registration can be both time consuming and inaccurate, and the change in alignment between preoperative imaging and intraoperative positioning further decreases accuracy.

Isocentric C-arm technology enhances spinal navigation by enabling the intraoperative acquisition of cone-beam CT images, making the process of image registration automatic (Fig. 1). Intraoperative CT imaging overcomes registration problems because the patient is already positioned on a radiolucent table when the scans are taken. Thus, tracking the position of the C-arm head as images are acquired enables fully automatic registration.

Control of radiation and accuracy
Radiation exposure to the patient, staff, and surgeon can be controlled, because intraoperative imaging can be used at the start of a procedure before surgical staff enter the field, or at the end after they have left the field. Tomography can be repeated at any time to verify registration accuracy and improve confidence in accurate placement. Verifying accuracy is especially important in longer procedures with multiple levels, in which the longer distance between fiducial markers and surgical site increases the likelihood of error.

Product availability
Currently, several major manufacturers offer isocentric systems with cone beam CT reconstruction, and others are developing systems. Current systems vary greatly in price, ease-of-use, size, maneuverability, and functionality (
Table 1).

Price is an essential factor in choosing a system. Several systems are available in the $300,000-to-$350,000 price range, while at least one system ranges from $650,000 to $800,000, depending on features.

The size of the system needs to be matched to the size of the target operating rooms, especially when a navigation system and other equipment will occupy the room. Systems with a smaller footprint can also serve as typical C-arms when CT reconstructions are not being used (Fig. 2).

Additional technical features that may affect purchasing decisions include imaging parameters and mechanical versus automatic positioning. Which additional features are desirable will be specific to each work environment.

Recent advances in integration technologies have increased compatibility among isocentric C-arm products and navigation products from different vendors.

Image quality and radiation exposure are complex issues, but all current commercial systems have high image quality and adjustable parameters to balance radiation dosage and maximize patient safety.

Impact on navigation
Increased use of isocentric C-arm technology has accelerated interest in spinal navigation, which has not been widely adopted despite its long history. The advent of intraoperative cone-beam CT allowed by isocentric C-arms represents a new, third paradigm in spinal navigation technology.
Figure 3 illustrates multiplanar navigation views from one commercially available system that can be automatically registered with the use of an isocentric C-arm.

In the past, spinal navigation could be performed in two ways. Originally, preoperative CT scans were obtained and intraoperative registration was done manually. This required a special preoperative CT protocol and cumbersome intraoperative registration. Subsequently, “virtual fluoroscopy” was developed in which two dimensional-fluoroscopy images were obtained and registered intraoperatively with a computer model. This eliminated the need for preoperative CT scans, but still required manual registration and did not allow axial tomographic views useful for spinal procedures.

Isocentric C-arm technology allows fully automatic registration based on intraoperative CT alone, retaining the advantages of both prior approaches. However, controlled clinical trials that demonstrate improved outcomes or improved patient safety with the use of spinal navigation techniques are lacking.

Currently, multiple vendors offer spinal navigation systems with trackers that integrate intraoperative isocentric C-arm data. Trackers are optical or electromagnetic and are typically placed in the bone of the ileum or spinous processes.

Trackerless navigation technology is also available from some vendors. Such systems rely on registration of preoperative magnetic resonance (MR) or CT images with intraoperative fluoroscopy images.

Both tracked and trackerless systems utilize frames that are placed on the collector of the C-arm and contain registration grips or accelerometers. Preoperative MR may be integrated into some systems.

The final analysis
As spinal navigation technology has evolved, surgeons have critiqued the time requirement and accuracy of manual registration (with or without trackers) and the need for preoperative imaging. Advantages of navigation may include reduced radiation to surgical staff and potentially improved accuracy and patient safety in some situations.

The increasing adoption of cone-beam CT allowed by isocentric C-arm technology may improve the risk/benefit profile of surgical navigation and contribute to the further evolution of both technologies. Adequately designed clinical trials to show safety and efficacy are required, and controlled trials are essential.

In all cases, the surgeon’s experience, tactile feedback, and multiple visual landmarks and cues must be used to ensure patient safety throughout a procedure. As with other imaging, the risks and benefits of radiation exposure must be managed to optimize patient safety.

S. Raymond Golish, MD, PhD, and Paul A. Anderson, MD, are members of the AAOS Biomedical Engineering Committee. Dr. Golish can be reached at rgolish@peacehealth.org; Dr. Anderson can be reached at anderson@ortho.wisc.edu