Microbubble Contrast Agent with Focused Ultrasound to Create Brain Lesions at Low Power Levels: MR Imaging and Histologic Study in Rabbits

Abstract

To evaluate magnetic resonance (MR) imaging-based thermometry for predicting the onset and spatial extent of lesions produced by focused ultrasound combined with a microbubble contrast agent (Optison; GE Healthcare, Milwaukee, Wis) and to compare the resulting induced temperature increase and threshold for damage with those in studies performed without the agent. The experiments were approved by the animal care committee. Fifty-three locations in the brains of 15 rabbits were sonicated with various exposure parameters by using a 1.5-MHz focused ultrasound transducer. MR imaging was used to map the temperature rise and, along with light microscopy, to examine the lesions. Diameters of isotherms created from thermometry were compared with the resulting lesions by using Bland-Altman analysis and linear regression. The minimum acoustic power necessary for lesion creation was determined, and the apparent temperature threshold for damage was calculated with probit analysis. These thresholds were compared with prior work performed without the contrast agent. The heating induced with the microbubbles was compared with that in sonications performed without them by using a t test. The MR imaging-mapped temperature distributions matched the shape of the lesions. The diameters of isotherms correlated well with diameters measured at contrast material-enhanced MR imaging (mean difference between measurements, 0.0 mm +/- 0.5; R = 0.93). The temperature increase with microbubbles was statistically larger (P < .01) than for sonications performed without microbubbles. In some locations (mostly continuous wave exposures), damage was observed along the ultrasound beam path. The time-averaged acoustic power damage threshold was reduced by 91% for 10-second exposures when compared with earlier studies performed without microbubbles. The probability of producing lesions was 50% at a temperature increase of 5.9 degrees C, 5.5 degrees C lower than was observed earlier without the agent. MR imaging-based temperature measurements appeared to correlate with focused ultrasound-induced lesions in the brain when microbubbles were present, even though the temperature appeared to be below the threshold for thermal damage.