Purpose: To investigate the influence of local density increase by i.v. contrast agent on dose calculation in linac-based radiosurgery (RS) of cerebral arteriovenous malformations (AVMs).
Material and methods: RS was performed after three-dimensional (3-D) treatment planning using a total number of nine to 14 beams. Mean target volume was 5.3 cm(3) (range, 0.1-41.2 cm(3)). Mean maximum diameter was 23.2 mm (range, 8-51 mm). Dose deviation was estimated and calculated from the enhanced and unenhanced datasets of 30 patients. Dose calculation was performed using the same RS treatment plan on both datasets. Both plans were standardized to 1 Gy at isocenter with the same dose weight for all beams.
Results: Mean difference of Hounsfield units (DeltaHU) between enhanced and unenhanced CT was 152 HU (range, 50-350 HU). The estimated dose deviation was <or= 1% in 80% of cases with a mean deviation of 0.67% and a maximum dose deviation of 1.8%. With increasing DeltaHU and increasing maximum diameter dose deviation increased as well. The calculated overdosage in ten datasets of enhanced and unenhanced CT scans was 0.66% mean (range, 0.2-1.2%).
Conclusion: The use of i.v. contrast agent in 3-D treatment planning for RS of cerebral AVMs may lead to an underestimation of actual applied dose. The effect on dose calculation is rather low with dose deviations < +1% in most of the cases. However, there are cases especially in large AVMs with high DeltaHU located next to critical, radiosensitive structures in which an additional unenhanced CT scan is recommended for exact dose calculation to avoid side effects.