In vitro and in vivo ablation of porcine renal tissues using high-intensity focused ultrasound

Abstract

The aim of this paper is to present issues regarding the thermal ablation of porcine renal tissues in vitro and in vivo using high-intensity focused ultrasound (HIFU). Production of lesions in the cortex in vitro is consistent, whereas lesions in the medulla are created whenever there are no air spaces in the medulla. Typically, the lesion length at 2000 W/cm² and 5-s pulse duration is around 20 mm and the corresponding width around 3 mm. Lesioning of a large volume was achieved by moving the transducer in a grid formation. Lesioning through a fat layer is possible provided that there are no air spaces between the fat and kidney interface. It was found that, above 3200 W/cm² with 5-s pulse duration at 4 MHz, cavitation activity occurred in most of the lesions created. (E-mail: [email protected])

Introduction

High-intensity focused ultrasound (HIFU) is a noninvasive procedure for heating tumours without affecting the healthy tissue surrounding the tumour. Therefore, application of HIFU is being investigated as an alternative to standard surgical techniques. Although the idea of using HIFU was proposed in the middle of this century by Lynn et al. (1942), its maximum potential for clinical use has been established only recently, due to the developments of sophisticated systems (for example, Chapelon et al., 1992b, Hynynen et al., 1993a and Birhle et al. 1994). HIFU has been proven over nearly 60 years to be an effective and efficient method for ablating soft tissue, but its commercial success is still under trial.

HIFU was explored in almost every tissue that is accessible by ultrasound. The following represent some examples of some applications explored: eye (Lizzi et al. 1984), prostate Sanghvi et al., 1991, Chapelon et al., 1999, liver (ter Haar et al. 1989), brain Fry et al., 1954, Lele, 1962, Vykhodtseva et al., 1994 and kidney Linke et al., 1973, Chapelon et al., 1992a, Hynynen et al., 1995.

Recently, the technology of HIFU systems has improved because the ultrasonic therapy can be guided by imaging. Ultrasonic imaging is the simplest and most inexpensive method; however, it has poor contrast between soft tissues. On the other hand, magnetic resonance imaging (MRI) offers superior contrast, but it is more expensive. Several trials have been conducted in the area of ultrasonic imaging (for example, Seip and Ebbini, 1995, Maass-Moreno et al., 1996 and in the area of MRI (for example, Cline et al., 1992, Hynynen et al., 1993a, Hynynen et al., 1995, Hynynen et al., 1993b, thus enhancing the potential of HIFU.

The main goal of HIFU is to maintain a temperature between 50 to 100°C for a few s (typically < 10 s), in order to cause tissue necrosis. Typically, focal peak intensity between 1000 to 10,000 W/cm² is used with pulse duration between 1 to 10 s and a frequency of 1 to 5 MHz.

This paper reports additional experience in the ablation of kidney tissue. For applications such as the treatment of benign prostatic hyperplasia (BPH), it is sufficient to destroy as much tissue as possible for the purpose of tissue debulking. For applications in oncology, destruction of all viable tumor cells is required and, therefore, protocols in this area must be very accurate and reliable. In this work, methods suggested by Malcolm and ter Haar (1996) were used to produce complete, reliable and consistent ablation of renal tissues (creation of a contiguous array of touching lesions of thermal origin, avoiding boiling and cavitation, and use of cooling to avoid merging of lesions in front of the focus).

Experiments were performed using a generic HIFU system that includes a signal generator, a radiofrequency (RF) amplifier, a 3-D robotic system, a transducer and a personal computer (PC) that controls the entire system. The transducer of 4-cm diameter was spherically focused, operating at 4 MHz. Pulse duration of 5 s was used in all the experiments to minimise effects of blood perfusion.

Results of HIFU ablation of porcine kidney in vitro and in vivo are presented. Single lesions in the cortex and medulla with or without a fat layer were created. The ultimate goal was to create large lesions, which was accomplished by moving the transducer in patterned schemes. The experimental results are compared with the results of a simulation model that includes the effect of rapid increase of attenuation during the transition of a tissue from healthy to necrotic. Thus, the kidney attenuations of porcine cortex, medulla, muscle and fat were measured as a function of thermal dose.

The attenuation was measured using a system that includes two low-intensity transducers, a signal generator, a data-acquisition card and a PC. Attenuation includes absorption, scattering and reflection but, when minimising scattering and reflection, attenuation will reflect mostly losses due to absorption. Scattering and reflection was minimised by using a homogeneous sample with a minimum number of interfaces or air ducts, and by removing acoustically troublesome gas bubbles using degassing techniques. Using these precautions in the study by Damianou et al. (1997), it was found that the value of absorption was very close to the value of attenuation, indicating that the contribution of reflection and scattering was minimised. Absorption is an acoustical property of tissue with wide variation from tissue to tissue and represents the rate at which energy in tissue is converted to heat. For short pulses (< 5 s), it is one of the primary factors contributing to the temperature elevation in tissue because the effect of blood flow is minimised.