Controlled ultrasound tissue erosion

Abstract

The ability of ultrasound to produce highly controlled tissue erosion was investigated. This study is motivated by the need to develop a noninvasive procedure to perforate the neonatal atrial septum as the first step in treatment of hypoplastic left heart syndrome. A total of 232 holes were generated in 40 pieces of excised porcine atrial wall by a 788 kHz single-element transducer. The effects of various parameters [e.g., pulse repetition frequency (PRF), pulse duration (PD), and gas content of liquid] on the erosion rate and energy efficiency were explored. An Isppa of 9000 W/cm2, PDs of 3, 6, 12, and 24 cycles; PRFs between 1.34 kHz and 66.7 kHz; and gas saturation of 40-55% and 79-85% were used. The results show that very short pulses delivered at certain PRFs could maximize the erosion rate and energy efficiency. We show that well-defined perforations can be precisely located in the atrial wall through the controlled ultrasound tissue erosion (CUTE) process. A preliminary in vivo experiment was conducted on a canine subject, and the atrial septum was perforated using CUTE.

Fig. 1

Experimental setup. Connection A with the 20-MHz pulser-receiver was used to measure the thickness of the tissue. Connection B, monitoring the 788 kHz echo, is used to indicate the time of tissue perforation.

Fig. 2

An example of the measurement of the tissue thickness. The 20-MHz transducer was pulsed and the echo received from the target tissue. Echo A was from the front surface of the tissue and echo B was from the back surface of the tissue. The time from echo A to B was 2.3 μs, resulting in a computed thickness of 1.8 mm. (The speed of sound in porcine cardiac tissue is 1.572 mm/μs at 26°C [26].)

Fig. 3

An example of using the 788 kHz echo detected by the 20-MHz transducer as a monitor of tissue perforation. Echo A was from the tissue, which disappeared (echo B) as the tissue was eroded through (PD = 3 cycles in the waveform and PRF = 13.9 kHz).

Fig. 4

The waveform of the therapeutic ultrasound signal with different PDs delivered by the 788-kHz therapy transducer as recorded by a membrane hydrophone. This figure shows the actual acoustic waveforms with complete ring up and down times. The PD used in this paper is defined as the number of cycles in the waveform at the output of the function generator except for the calculation for Isppa. Panels A, B, C, and D are waveforms in the pressure field at the focus with PDs of 3, 6, 12, and 24 cycles, respectively.

Fig. 5

A well-defined hole was created in the porcine atrial wall with a PD of 3 cycles and a PRF of 19.6 kHz.

Fig. 6

The relationship between the erosion rate and PRF for a fixed PD. In panels A, B, C, and D, the erosion rate was plotted versus PRF by median and range values (min. and max.) for N = 8, corresponding to PDs of 3, 6, 12, and 24 cycles, respectively. In panel A, B, and C, for PDs of 3, 6, and 12 cycles, there is a tendency for the erosion to be faster at PRFs of 19.6 kHz, 6.90 kHz, and 4.95 kHz, respectively. In panel D, for a PD of 24 cycles, the erosion rate increases with the increasing PRF.

Fig. 7

The relationship between the energy efficiency and PRF for a fixed PD. In panels A, B, C, and D, the energy efficiency was plotted versus PRF by median and range values (min. and max.) for N = 8, corresponding to PDs of 3, 6, 12, and 24 cycles, respectively. In panels A and B, for PDs of 3 and 6 cycles, there are local peaks of energy efficiency at PRFs of 13.9 kHz and 6.90 kHz, respectively. In panels C and D, local peaks of energy efficiency are not distinct.

Fig. 8

The relationship between the erosion rate and PD for a fixed duty cycle. In panels A, B, C, and D, the erosion rate was plotted versus PD by median and range values (min. and max.) for N = 8, corresponding to duty cycles of 5%, 7.5%, 11%, and 25%, respectively. In panel A, with a 5% duty cycle, the median erosion rate is 10.54 μm/s for a PD of 3 cycles and 1.23 μm/s for a PD of 24 cycles, approximately an 8-fold difference. In panel B, with a 7.5% duty cycle, the median erosion rate is 12.10 μm/s for a PD of 3 cycles and 2.01 μm/s for a PD of 24 cycles, approximately a 6-fold difference.

Fig. 9

The relationship between the energy efficiency and PD for a fixed duty cycle. In panels A, B, C, and D, the energy efficiency was plotted versus PD by median and range values (min. and max.) for N = 8, corresponding to duty cycles of 5%, 7.5%, 11%, and 25%, respectively. In panel A, with a duty cycle of 5%, the energy efficiency is 0.99 μm/J for a PD of 3 cycles and 0.12 μm/J for a PD of 24 cycles, approximately an 8-fold difference. In panel B, with a duty cycle of 7.5%, the energy efficiency is 0.81 μm/J for a PD of 3 cycles and 0.14 μm/J for a PD of 24 cycles, approximately a 6-fold difference.

Fig. 10

The relationship between the erosion rate and gas saturation. Panel A is the plot of erosion rate versus PRF, with 40–55% gas saturation and a PD of 3 cycles. Panel B is the plot of erosion rate versus PRF with 79–85% gas saturation and a PD of 3 cycles. The erosion rate was plotted by median and range values (min. and max.) for N = 8. In panel A, the median erosion rate is 12.10 μm/s with a PRF of 19.6 kHz and gas saturation of 40–55%. In panel B, the median erosion rate is 3.91 μm/s with a PRF of 19.6 kHz and gas saturation of 79–85%.

Fig. 11

The relationship between the energy efficiency and gas saturation. Panel A is the plot of energy efficiency versus PRF, with 40–55% gas saturation and a PD of 3 cycles. Panel B is the plot of energy efficiency versus PRF with 79–85% gas saturation and a PD of 3 cycles. The energy efficiency was plotted by median and range values (min. and max.) for N = 8. In panel A, the median energy efficiency is 0.99 μm/J with a PRF of 13.9 kHz and gas saturation of 40–55%. In panel B, the median erosion rate is 0.30 μm/s with a PRF of 13.9 kHz and gas saturation of 79–85%.

Fig. 12

The picture on the left shows holes created with a PD of 3 cycles and 40% gas saturation. The picture on the right depicts the holes created with a PD of 24 cycles and 55% gas saturation. The diagrams depict the formation of the holes corresponding to the pictures above them.

Fig. 13

In vivo experiment. Panel A shows the Doppler ultrasound image of the blood flow resulting from the perforation of the atrial septum between the two atria (arrows). Panel B shows the existence of the perforation of the atrial septum by inserting a blunt gauge 18 needle (~1 mm in diameter) through the hole.