JOURNAL OF SURGICAL RESEARCH 71, 19–24 (1997) ARTICLE NO. JR975133 Ex Vivo Experiment on Radiofrequency Liver Ablation with Saline Infusion through a Screw-Tip Cannulated Electrode Yi Miao, M.D.,* Yicheng Ni, M.D.,* Stefaan Mulier, M.D.,† Kai Wang, M.D.,‡ Michael F. Hoey, M.D.,§ Peter Mulier, Ø Freddy Penninckx, M.D.,† Jie Yu, M.D.,* Ivan De Scheerder, M.D.,‡ Alber L. Baert, M.D.,* and Guy Marchal, M.D.* *Department of Radiology, †Department of Abdominal Surgery, and ‡Department of Cardiology, University Hospitals, Leuven, B-3000 Belgium; and §Department of Physiology, University of Minnesota, and Ø Medtronic Inc., Minneapolis, Minnesota Submitted December 26, 1996 saline into the tissue through the hollow electrode tip Purpose: To investigate whether radiofrequency [6–8]. The conductivity of normal saline is 3 to 5 times (RF) therapy with hypertonic saline infusion through higher than that of the blood and 12 to 15 times higher a hollow screw-tip electrode can cause a lesion size than that of tissues. Infused saline produces a liquid suitable for liver tumor ablation. Materials and meth- electrode in the tissue to be ablated and spreads the ods: RF tissue ablation of 180 sites was performed by applied RF energy away from the metal electrode. The using a hollow screw-tip electrode in 40 freshly excised resultant lower current density at the metal electrode – swine livers. Under both power and temperature con- tissue interface reduces tissue desiccation [6]. When trol modes, the ablation effects with and without vari- room temperature saline is infused, some convective ous regimes of 5% hypertonic saline (1 ml/min) prior cooling also occurs at the tip. The purpose of this study to and/or during the procedure were compared by was to evaluate whether the RF ablation size of liver measuring the size of lesions at dissection and con- tissue (eventually in liver tumors) can still be increased firmed by T1 and T2 weighted magnetic resonance by interstitial infusion of 5% hypertonic saline through (MR) imaging. Results: The maximal lesion diameter a newly developed screw-tip hollow electrode (Med- of 5.5 cm was reached at 30 W with saline infusion 1 tronic, Minneapolis, MN). min prior to and during 12 min of ablation. The smaller sizes (P õ 0.01) between 0.3 and 2.5 cm in diameter were met with noninfusion or preinfusion-only groups. MATERIALS AND METHODS The RF ablation lesions appeared as hyper- and hypo- Ablation protocol. RF tissue ablation of 180 sites was performed intense areas on T1 and T2 MR images, respectively. in 40 freshly excised swine livers. Four to 12 applications of RF Conclusions: RF ablation in combination with present current were delivered to separate sites per excised liver. RF current hollow screw-tip electrode and saline infusion allows was produced by a modified RF generator (Medtronic, San Jose, CA); for necrotic development of suitable size for liver tu- the maximal energy output is 50 W into a 50–250 V load of 475 { mor ablation. Such ablated lesions can be visualized 20 kHz RF current continuously with an automatic high impedance with MR imaging.  1997 Academic Press shutoff at 250 V. Impedance, tip electrode temperature, and power output values are recorded along with the time of ablation. A luminal screw catheter (5 French) with one end hole and four side holes (0.12 mm in diameter) at its tip, which is 1.7 mm in screw diameter and INTRODUCTION 10 mm in length (Fig. 1), was inserted into the tissue at a depth of 1.5 to 3 cm. This catheter permits infusion of fluid and delivery of RF current. RF current was delivered between the catheter tip elec- Radiofrequency (RF) catheter ablation techniques trode and an adhesive electrosurgical pad placed on the opposite have shown promise for treatment of either primary or side of the liver. The delivery was controlled using both power and metastatic solid neoplasms in the liver [1, 2]. However, temperature mode. An additional thermocouple (Type K, Omega En- the maximum size of a lesion that can be treated effec- gineering, Stamford, CT) was inserted in contact with the distal electrode to monitor tip temperature. In the power control mode, 30 tively is still limited by impedance rise from the elec- W was chosen because of the match between power and saline infu- trode – tissue interface, secondary to tissue desiccation sion rate (1 ml/min). In the temperature control mode, the RF genera- and charring [1, 2]. A proposed approach to solve this tor output was adjusted to the greatest level of power without in- limitation is to irrigate the electrode with cooling liquid creasing electrode temperature beyond 90°C. Hypertonic saline (5% [3–5]. By the cooling effect, the tip of the electrode is NaCl, Baxter N.V., Lessines, Belgium) at room temperature (20°C) was infused at 1 ml/min by using an infusion pump (Ismatic, Switzer- maintained at low temperature, thereby increasing the land). The experimental regimes were designed as both power and conductivity of the electrode – tissue interface and pre- temperature control modes. Each mode was further divided into venting an impedance rise. The effectiveness of this three subgroups: approach has been shown in the ablation of abnormal Power control mode (energy output fixed at 30 W): Group A (Ap), no infusion, n Å 10; Group B (Bp), 1 min preinfusion only (1 ml), n muscle and liver [3–5]. Another approach is to infuse 19 0022-4804/97 $25.00 Copyright  1997 by Academic Press All rights of reproduction in any form reserved. AID JSR 5133 / 6n21$$$101 07-14-97 16:57:49 srga