Microbubble-enhanced transcranial MR-guided focused ultrasound brain hyperthermia: heating mechanism investigation using finite element method

Abstract

Recently, our group developed a synergistic brain drug delivery method to achieve simultaneous transcranial hyperthermia and localized blood-brain barrier opening via MR-guided focused ultrasound (MRgFUS). In a rodent model, we demonstrated that the ultrasound power required for transcranial MRgFUS hyperthermia was significantly reduced by injecting microbubbles (MBs). However, the specific mechanisms underlying the power reduction caused by MBs remain unclear. The present study aims to elucidate the mechanisms of MB-enhanced transcranial MRgFUS hyperthermia through numerical studies using the finite element method. The microbubble acoustic emission (MAE) and the viscous dissipation (VD) were hypothesized to be the specific mechanisms. Acoustic wave propagation was used to model the FUS propagation in the brain tissue, and a bubble dynamics equation for describing the dynamics of MBs with small shell thickness was used to model the MB oscillation under FUS exposures. A modified bioheat transfer equation was used to model the temperature in the rodent brain with different heat sources. A theoretical model was used to estimate the bubble shell's surface tension, elasticity, and viscosity losses. The simulation reveals that MAE and VD caused a 40.5% and 52.3% additional temperature rise, respectively. Compared with FUS only, MBs caused a 64.0% temperature increase, which is consistent with our previous animal experiments. Our investigation showed that MAE and VD are the main mechanisms of MB-enhanced transcranial MRgFUS hyperthermia.

Keywords: Bubble dynamics; Finite element method; MR-guided focused ultrasound; Microbubble; Transcranial hyperthermia.

Fig. 1

Schematic diagram of microbubble-enhanced transcranial MR-guided focused ultrasound brain hyperthermia, which can achieve simultaneous mild heating and blood–brain barrier opening for enhanced brain drug delivery.

Fig. 2

Cross-section of the 3D numerical model built in COMSOL. The spherical transducer with a diameter of $25\mathrm{m}\mathrm{m}$ and $f$-number of $0.8$ was placed at the bottom of the model. A water layer was placed between the transducer and brain tissue. MBs were randomly distributed in the ultrasound focal region (the ellipse). The outermost layer of the model is the perfect matching layer. The red area with dots indicates the volume where the bioheat transfer calculation was implemented. The temperature boundary condition along the purple dashed line for bioheat transfer calculation was set to be $37$ °C.

Fig. 3

The acoustic pressure distribution of the simulated acoustic field. The absolute acoustic pressure distribution at YZ-plane ($X=0\mathrm{m}\mathrm{m}$): A) with FUS only, B) with FUS + MBs. The profile of the acoustic pressure amplitude with and without MBs: C) along the FUS beam, D) across the center of the FUS beam.

Fig. 4

Representative microbubble behaviors in the simulation under FUS exposures from four bubbles. A-D) ultrasound pressure at the bubble location as a function of time; E-H) relative radius change as a function of time; I-L) the pressure of MAE at $10R_0$ from the bubble center as a function of time; M−P) the spectra of MAE at the location of 10 $R_0$ from $t=0\mathrm{\mu }\mathrm{s}$ to $100\mathrm{\mu }\mathrm{s}$.

Fig. 5

Microbubble distribution and the temperature distribution with different heat sources. A) shows the MB distribution in this study. The spheres indicate the MBs while the red dots show the projection position of MBs on the YZ-plane, XZ-plane, and XY-plane. The semitransparent blue ellipsoid illustrates the focal region. B-E illustrate the temperature maps with four heating sources at YZ-plane ($X=0\mathrm{m}\mathrm{m}$): B) FUS only, C) FUS + MAE, D) FUS + VD, and E) FUS + MAE + VD. The cross signs in the inset of E) indicate MB positions at the YZ-plane. F) indicates the temperature rise profile across the center of the FUS beam axially with $\left(X,Y\right)=\left(0,0\right)\mathrm{m}\mathrm{m}$. G) the axial temperature rise profile with $\left(X,Y\right)=\left(0.2,0\right)\mathrm{m}\mathrm{m}$. H) shows the average temperature rise in the focal region as a function of time with four heating sources.