Setup accuracy of spine radiosurgery using cone beam computed tomography image guidance in patients with spinal implants

Abstract

Object Cone beam computed tomography (CBCT) image guidance technology has been adopted for use in spine radiosurgery. There is concern regarding the ability to safely and accurately perform spine radiosurgery without the use of implanted fiducials for image guidance in postsurgical cases in which titanium instrumentation and/or methylmethacrylate (MMA) has been implanted. In this study the authors prospectively evaluated the accuracy of the patient setup for spine radiosurgery by using CBCT image guidance in the context of orthopedic hardware at the site of disease. Methods The positioning deviations of 31 single-fraction spine radiosurgery treatments in patients with spinal implants were prospectively evaluated using the Elekta Synergy S 6-MV linear accelerator with a beam modulator and CBCT image guidance combined with a robotic couch that allows positioning correction in 3 translational and 3 rotational directions. To measure patient movement, 3 quality-assurance CBCT studies were performed and recorded: before, halfway through, and after radiosurgical treatment. The positioning data and fused images of planning CTs and CBCTs from the treatments were analyzed to determine intrafractional patient movements. From each of 3 CBCTs, 3 translational and 3 rotational coordinates were obtained. Results The prescribed dose to the gross tumor volume for the cohort was 12–18 Gy (mean 14 Gy) utilizing 9–14 coplanar intensity-modulated radiation therapy (IMRT) beams (mean 10 beams). At the halfway point of the radiosurgery, the translational variations and standard deviations were 0.6 ± 0.6, 0.4 ± 0.4, and 0.5 ± 0.5 mm in the lateral (X), longitudinal (Y), and anteroposterior (Z) directions, respectively. The magnitude of the 3D vector (X,Y,Z) was 1.1 ± 0.7 mm. Similarly, the variations immediately after treatment were 0.5 ± 0.3, 0.4 ± 0.4, and 0.5 ± 0.6 mm along the X, Y, and Z directions, respectively. The 3D vector was 1.0 ± 0.6 mm. The mean rotational angles were 0.3 ± 0.4, 0.5 ± 0.6, and 0.3 ± 0.4° along yaw, roll, and pitch, respectively, at the halfway point and 0.3 ± 0.4, 0.6 ± 0.6, and 0.4 ± 0.5° immediately after treatment. Conclusions Cone beam CT image guidance used for patient setup for spine radiosurgery was highly accurate despite the presence of spinal instrumentation and/or MMA at the level of the target volume. The presence of such spinal implants does not preclude safe treatment via spine radiosurgery in these patients.