Plain Film Radiography

Plain film radiography was the first imaging technique applied to spine surgery. The segmented bony anatomy of the spinal column lends itself readily to this form of imaging. Images are easily obtained, relatively inexpensive, and generally reliable. Differential attenuation of x-rays by the various tissues provides the image contrast seen in a radiograph. Bone attenuates x-rays the most, followed by muscle, fat, and air, which attenuate very few x-rays. Soft tissue structures such as the spinal cord, nerve roots, or intervertebral discs are not well visualized by plain film radiography.

In spine surgery, plain radiography is typically used to localize a specific spinal level. During the surgical exposure of the cervical, lumbar, or thoracic spine, a radiopaque marker (e.g., clamp, probe, spinal needle) is positioned in the surgical field. For the lumbar or cervical spine, a lateral radiograph is obtained with the lumbosacral or occipitocervical junction at one end and the inserted marker at the other end. The number of vertebral levels between the imaged junction and the marker can be determined and the appropriate spinal level easily localized. In general, the range of view from the occipitocervical junction extends caudally to the cervicothoracic junction, depending on the prominence of the patient’s neck and shoulders. The range of view from the lumbosacral junction extends rostrally to the lower thoracic region.

Spinal levels in the mid- and upper thoracic spine can be more difficult to precisely localize because of their distance from a reliable anatomic reference point. In this case, an alternative to the conventional method of radiographic localization is to obtain two adjacent lateral images. The first film is positioned to include the lumbosacral or occipitocervical junction at one end and an instrument marker attached to the spinal anatomy at the other end. A second film can then be positioned to include the first instrument marker and a second instrument marker centered in the operative field. The location of the first instrument marker can be determined on the first film by its relationship to the lumbosacral or occipitocervical junction. The location of the appropriate spinal level can then be determined on the second film by determining its relationship to the localized first instrument marker.

Radiographic imaging of the upper thoracic spine can be difficult because the patient’s shoulders may obstruct the view on a lateral radiograph. An alternative to the lateral radiograph is the anteroposterior (AP) view. This view can be obtained by positioning the patient on a radiolucent operating table. The film cassette can then be placed immediately below the table and an AP view obtained. With this view, either the T1 or T12 vertebra and its associated ribs serve as the reference point for spinal level identification.

When intraoperative localization involves exposure of the spinal column at the level of a collapsed or fractured vertebra, the relationship of the operative field to the lumbosacral or occipitocervical junction is not as critical. A film that centers over the operative field can approximate the appropriate level. The abnormal vertebrae can usually be identified on a lateral radiograph and the appropriate exposure confirmed.

The most common error related to the use of intraoperative plain radiography is incorrectly identifying the appropriate spinal level. This can be due to inadequate image quality that obscures the appropriate spinal anatomy. This can lead the surgeon to misinterpret the radiograph and incorrectly count the number of spinal levels from the lumbosacral or occipitocervical junction to the operative field. Suboptimal imaging should not be accepted. If satisfactory localization cannot be obtained, the radiograph should be repeated until correct localization is confirmed.

Even when the correct level is identified, localization errors may occur. For example, localization in the lumbar spine typically involves placing a marker on a spinous process and obtaining a lateral radiograph. Although the correct spinous process may be visualized on the film, continued surgical dissection down to the level of the lumbar canal may cause the surgeon to easily drift to an adjacent lumbar segment above or below the desired level. Placement of a marker at the level of the lumbar lamina in addition to the marker on the spinous process can help prevent this error.

In addition to localization of the spinal level, plain radiography can be used in isolated instances to assess the extent of neural decompression. This technique is particularly useful during transoral decompressive procedures. It involves placing a radiopaque contrast medium into the decompressed site and obtaining a lateral radiograph. The radiograph will show the extent of the decompression and can be compared with a preoperative study demonstrating the neural compression. If the configuration and location of the contained contrast medium does not approximate the configuration and location of the epidural compression on the preoperative studies, the surgical site can be modified until a satisfactory decompression of the neural elements is achieved (Fig. 175-1).

FIGURE 175-1 Lateral radiograph after a transoral odontoidectomy. Radiopaque contrast media has been placed into the decompressed site. The position and configuration of the contrast media can be compared to the preoperative studies to confirm a satisfactory decompression.

Plain radiography can also be used in conjunction with intraoperative myelography to determine the extent of canal decompression.¹ In the setting of a thoracolumbar burst fracture decompressed through a dorsolateral approach, radiopaque dye can be placed in the subarachnoid space by a small-gauge spinal needle to assess the adequacy of the surgical decompression. A lateral radiograph is taken and the relationship of the dye to the ventral spinal cord can be assessed (Fig. 175-2).

FIGURE 175-2 Intraoperative lateral myelography after lumbar decompression and fixation.

Fluoroscopy

Fluoroscopy is used when real-time imaging or multiple images of the spine are required during surgery. Fluoroscopic images are generated with a lower radiation dose than that for standard radiography. A camera records the image and displays it on a television monitor. The images can be recorded on x-ray film, cinefluoradiography film, or videotape.²

The primary use of fluoroscopy in spine surgery is to facilitate the optimal placement and positioning of spine fixation screws, interbody cages, or arthroplasty devices and to monitor the injection of methyl methacrylate into osteoporotic vertebral body fractures. Fluoroscopy can provide the surgeon with real-time imaging of the spinal column in several planes depending on the positioning of the fluoroscope’s C-arm. When used for the insertion of pedicle screws, the C-arm can be positioned to provide a lateral view of the spinal column. This shows the sagittal (rostrocaudal) orientation of a pedicle marker or screw (Fig. 175-3). By rotating the C-arm from the lateral position toward the AP position, an oblique view of the spinal column can be obtained. This view demonstrates the position of a pedicle marker or screw in relationship to a cross-sectional view of each pedicle. If the correct entry point and trajectory in both the sagittal and axial planes has been properly selected, the oblique fluoroscopic view will demonstrate these markers within the cortical margins of the individual pedicles (Fig. 175-4). When the correct entry point and trajectories have been confirmed, the markers can be removed and the pedicle screws inserted along the same path.

FIGURE 175-3 Lateral fluoroscopic view of the lower lumbar spine showing the sagittal orientation of Steinmann pins placed within the pedicles of L3, L4, and L5.

FIGURE 175-4 Oblique fluoroscopic view of the lumbar spine showing the pedicles in cross section. Three Steinmann pins have been placed within the pedicles and are noted to be lying within the cortical margins of the pedicles.

When used for cervical screw fixation (e.g., ventral odontoid screw fixation), an AP as well as a lateral view may be required. This can be obtained by alternating the position of a single C-arm unit from the lateral to the AP position or by using two separate fluoroscopy machines with arms positioned 90 degrees with respect to each other.³

Fluoroscopy has also been used for ventral cervical plate instrumentation procedures to assist in the positioning and placement of the fixation plates and screws.⁴ However, the lower cervical and upper thoracic regions can frequently be difficult to visualize because of image obstruction by the patient’s shoulders. This problem can be addressed by placing an interscapular roll beneath the patient and taping the shoulders down to the table or by using wrist slings to pull down on the patient’s arms during imaging.

A potential problem with fluoroscopic imaging is difficulty in obtaining a true lateral image of the spinal anatomy. This can produce a parallax effect that can result in the suboptimal positioning of fixation screws. In the cervical spine, this problem can be addressed by ensuring that the angle of one mandible viewed on the fluoroscopy screen overlaps the angle of the other mandible. In the thoracic and lumbosacral spine, the superior and inferior cortical margins of the vertebrae can be used to ensure a true lateral image. If two cortical margins are seen at one end of a vertebral body, a true lateral image has not been obtained and an adjustment of the C-arm is needed.

The recent development of minimally invasive techniques for spine surgery has led to a significantly increased use of intraoperative fluoroscopy. The limited exposure of soft tissue in these procedures reduces the ability of the surgeon to directly visualize the pertinent surgical anatomy. This makes fluoroscopy a critical component of any minimally invasive procedure.

The primary disadvantage of intraoperative fluoroscopy is the exposure of the surgical team to repeated amounts of low-dose ionizing radiation. Several studies have attempted to quantify the amount of occupational radiation exposure to health-care professionals under different and real clinical scenarios.^5–9 These studies have typically demonstrated that the radiation exposure to hands, eyes, head and neck, and body during fluoroscopically assisted procedures is well below the recommended values for annual allowable occupational radiation exposure as outlined by the International Commission on Radiological Protection.¹⁰ However, the health-related risks of this chronic exposure to radiation remains relatively unknown and may not be realized for years.

Unlike other fluoroscopically assisted musculoskeletal procedures (e.g., intramedullary femoral nailing), the techniques and instruments used for pedicle fixation bring the surgeon’s hands very close to the primary area of exposure.^11,12 The amount of radiation exposure needed to achieve adequate visualization of the lumbar spine is greater than that for other anatomic sites.¹³ Rampersaud et al. reported on the radiation exposure occurring during the placement of thoracolumbar pedicle screws.¹⁴ Pedicle screws were placed into six cadaver specimens using fluoroscopy. Dosimeter badges and rings were placed on the surgeon’s neck, torso, and dominant hand. Radiation exposure to these sites was quantified.

Ninety-six pedicle screws were placed with an average of 8.5 fluoroscopic images taken per screw. The average exposure time per screw was 9.3 seconds. The results of this study demonstrated that fluoroscopically assisted pedicle screw fixation exposes the spine surgeon to significantly greater radiation levels than other, nonspinal musculoskeletal procedures involving the use of fluoroscopy. The radiation exposure to the dominant hand was noted to be as high as 10 to 12 times the amount of hand exposure experienced in other procedures. The greatest level of exposure was noted on the side ipsilateral to the beam source due to backscatter radiation. This study emphasizes the need for appropriate lead shielding, including the use of a thyroid shield and leaded glasses. The use of radiation attenuation gloves resulted in a 33% decrease in dose rate to the hand. Using limited, pulsed-image acquisition and maintaining a safe distance (3–4 feet) from the radiation source also helped reduce the dosage to negligible levels.¹⁴ This study confirmed that spine surgeons who regularly use fluoroscopy should monitor their annual radiation dose.

Ultrasonography

The application of ultrasonography to spine surgery was first reported in 1982.¹⁵ Like fluoroscopy, ultrasonography provides continuous and real-time imaging. However, unlike fluoroscopy, which only provides images of the bony spinal column, ultrasonography can image the adjacent neural and soft tissue structures as well.