The aim of this study was to test the hypotheses that (a) MR imaging-guided
radiofrequency (RF) thermal ablation is safe and feasible in porcine brain
using an open C-arm-shaped low-field MR system, and that (b) induced-therm
al lesion size can be predicted using low-field MR imaging. Magnetic resona
nce-guided RF ablation was performed in the cerebral frontal lobes of six p
igs. An 18-G monopolar RF electrode was inserted into the porcine brain usi
ng MR image guidance and RF was then applied for 10 min. After post-procedu
re imaging (T2-weighted, T1-weighted before and after gadodiamide administr
ation), the pigs were killed and the brains were used for pathologic examin
ation. Successful RF electrode placement was accomplished in all cases with
out complications; total magnet time ranged from 73 to 189 min. The thermal
lesion size varied from 10 to 12 mm perpendicular to the electrode track a
nd was easily visualized on T2-weighted and enhanced T1-weighted images. En
hanced T1-weighted imaging demonstrated the highest brain-to-RF thermal les
ion contrast-to-noise ratio with an average of 1.5 +/- 1.6. Enhanced T1-wei
ghted imaging never underestimated pathologic lesion diameter with a mean d
ifference of 2.3 +/- 1.0 mm and a radiologic/pathologic correlation of 0.69
, Magnetic resonance imaging-guided RF thermal ablation is feasible and saf
e in the porcine brain using an open MR low-field system. Induced thermal l
esion size can best be monitored using enhanced T1-weighted images. In the
future, RF ablation under low-field MR guidance may offer an alternative tr
eatment option for primary and secondary brain tumors.