Abstract
Ultrasound and microbubble enhanced drug delivery presents an opportunity to enhance therapy and improve patient outcomes. In addition to the enhancement of drug and gene uptake, ultrasound and microbubbles provide targeted drug and gene delivery, permeabilizing the cell membrane within the ultrasound beam width. A variety of indications can benefit from localized therapeutic delivery, especially in the vasculature. In particular, atherosclerosis, the build up of fat and cellular waste in the vasculature, is a viable target for a catheter-based ultrasound and microbubble enhanced drug delivery platform. Catheter procedures are already common tools for diagnosing and treating atherosclerosis through coronary angiography, intravascular ultrasound (IVUS) imaging and percutaneous coronary intervention (PCI).
This dissertation develops an IVUS platform for microbubble-based drug delivery as well as metrics for assessing microbubble delivery with IVUS. IVUS transducers for microbubble-based drug delivery were designed using finite element analysis (FEA) and an experimentally verified microbubble displacement model. The designed transducers were fabricated, characterized, and evaluated using an in vitro cell delivery model, flow phantoms, and ex vivo artery experiments.
An experimentally verified 1-D acoustic radiation force (ARF) microbubble model was implemented and evaluated for the selection of the microbubble-specific ultrasound transducer center frequencies. Outputs from the model were later compared to high speed camera data to determine the validity of the model as a tool for selecting transducer center frequency. Through thorough review of the equations which the model was based on and comparison with published results, a correction was made to the friction term of the model. After applying this correction, the impact of this friction term on the model output was evaluated.
Ultrasound transducer design frequencies were selected to displace microbubbles using the results of the ARF microbubble model and to induce sonoporation, the transient permeabilization of the cell membrane, using data found in the literature. Using FEA, low center frequency IVUS transducers (1.5, 2, and 5 MHz) that fit within the dimensional constraints presented by the vasculature (diameter < 1 mm, length < 3.5 mm ) were designed. Prototypes were fabricated using commercially available piezoelectric ceramics and characterized using a hydrophone measurement system. The measured prototype characteristics were compared to the FEA results. The measured and FEA results were used to assess the accuracy of the finite element model (FEM) and the FEM was adjusted to better match the measured results. The ability of these prototype transducers to displace microbubbles from physiological flow conditions was validated in flow phantoms while monitoring using both a clinical ultrasound scanner and programmable research ultrasound scanners.
To achieve a displace, treat, and image model of delivery, multifunction IVUS transducers for microbubble displacement, ultrasound and microbubble enhanced drug delivery, and imaging were designed using FEA. Using lateral and thickness modes, designs that incorporate both imaging center frequencies (8-24 MHz) and therapeutic center frequencies (< 5 MHz) were produced. Two of these designs were fabricated, characterized, and tested.
The ability of the IVUS transducers to induce sonoporation was evaluated in vitro. The dual frequency, multifunction IVUS transducers induced ultrasound and microbubble enhanced uptake of a membrane impermeable fluorophore (calcein) at a peak negative pressure (PNP) = 152 kPa and pulse repetition frequency (PRF) = 1 kHz, while maintaining vascular smooth muscle cell viability in in vitro studies.
Delivery (5 MHz center frequency, 50 cycle sine, PNP = 1 or 2 MPa, PRF = 0.5 or 1 kHz) and ARF (5 MHz center frequency, 500 cycle sine, PNP = 0.6 MPa, PRF = 5 kHz ) parameters were tested in an ex vivo swine artery flow loop using physiological flow rates (105 mL/min). The percentage of time dedicated to ARF and delivery was varied to determine which pulse had the greatest impact on the delivery of a model drug (DiI). It was determined that without ARF, delivery does not occur. However, without high amplitude sonoporation pulses, fluorescence intensities similar to a combination of ARF and delivery pulses were measured.
Finally, the ultrasound parameters determined from the ex vivo and in vitro experiments were applied to perform an in vivo pilot study in a swine model. DiI microbubbles were infused through an IVUS catheter as the transducer rotated and transmitted a 5 MHz center frequency, 500 cycle, PNP = 0.6 MPa, PRF = 5 kHz pulse to displace microbubbles to the vessel wall. Localized delivery of DiI was verified in the swine model through fluorescence microscopy of the swine model's left circumflex and left anterior descending arteries.
