Substantial dose reduction in modern multi-slice spiral computed tomography (MSCT)-guided craniofacial and skull base surgery

Abstract

Purpose: Reduction of the radiation exposure involved in image-guided craniofacial and skull base surgery is an important goal. The purpose was to evaluate the influence of low-dose protocols in modern multi-slice spiral computed tomography (MSCT) on target registration errors (TREs).

Materials and methods: An anthropomorphic skull phantom with target markers at the craniofacial bone and the anterior skull base was scanned in Sensation Open (40-slice), LightSpeed VCT (64-slice) and Definition Flash (128-slice). Identical baseline protocols (BP) at 120 kV/100 mAs were compared to the following low-dose protocols (LD) in care dose/dose modulation: (LD-I) 100 kV/35ref. mAs, (LD-II) 80 kV/40 - 41ref. mAs, and (LD-III) 80 kV/15 - 17ref. mAs. CTDIvol and DLP were obtained. TREs using an optical navigation system were calculated for all scanners and protocols. Results were statistically analyzed in SPSS and compared for significant differences (p ≤ 0.05).

Results: CTDIvol for the Sensation Open/LightSpeed VCT/Definition Flash showed: (BP) 22.24 /32.48 /14.32 mGy; (LD-I) 4.61 /3.52 /1.62 mGy; (LD-II) 3.15 /2.01 /0.87 mGy; and (LD-III) na/0.76 /0.76 mGy. Differences between the BfS (Bundesamt für Strahlenschutz) reference CTDIvol of 9 mGy and the lowest CTDIvol were approximately 3-fold for Sensation Open, and 12-fold for the LightSpeed VCT and Definition Flash. A total of 33 registrations and 297 TRE measurements were performed. In all MSCT scanners, the TREs did not significantly differ between the low-dose and the baseline protocols.

Conclusion: Low-dose protocols in modern MSCT provided substantial dose reductions without significant influence on TRE and should be strongly considered in image-guided surgery.