Abstract
Study design
Retrospective study.
Objectives
To optimise the radiation doses and image quality for the cone-beam O-arm surgical imaging system in spinal surgery.
Summary of Background
Neurovascular compromise has been reported following screw misplacement during thoracic pedicle screw insertion. The use of O-arm with or without navigation system during spinal surgery has been shown to lower the rate of screw misplacement. The main drawback of such imaging surgical systems is the high radiation exposure.
Methods
Chest phantom and cadaveric pig spine were examined on the O-arm with different scan settings: two were recommended by the O-arm manufacturer (120 kV/320 mAs, and 120 kV/128 mAs), and three low-dose settings (80 kV/80 mAs, 80 kV/40 mAs, and 60 kV/40 mAs). The radiation doses were estimated by Monte Carlo calculations. Objective evaluation of image quality included interobserver agreement in the measurement of pedicular width in chest phantom and assessment of screw placement in cadaveric pig spine.
Results
The effective dose/cm for 120 kV/320 mAs-scan was 13, 26, and 69 times higher than those delivered with 80 kV/80 mAs, 80 kV/40 mAs, and 60 kV/40 mAs-scans, respectively. Images with 60 kV/40 mAs were unreliable. Images with 80 kV/80 mAs were considered reliable with good interobserver agreement when measuring the pedicular width (random error 0.38 mm and intraclass correlation coefficient 0.979) and almost perfect agreement when evaluating the screw placement (κ-value 0.86).
Conclusions
The radiation doses of the O-arm system can be reduced 5–13 times without negative impact on image quality with regard to information required for spinal surgery.
