All animal experiments were performed in the Preclinical Animal Laboratory of Kagawa University (Kagawa, Japan). The protocol of this study was approved by the Institutional Animal Care and Use Committee of Kagawa University. Adult pig livers were purchased from Tokyo Shibaura Zoki K.K. (Tokyo, Japan). The 13-month-old dogs were purchased from Hokuzan Rabesu (Nagano, Japan) and maintained in a pathogen-free facility under controlled conditions with a 12-h light-dark cycle.

RFA and D-sorbitol RFA were performed under general anesthesia with intramuscular injection of ketamine (10 mg/kg) and xylazine hydrochloride (4 mg/kg). Intubation was performed 10–15 min later and the administration of 5% sevoflurane mixed with oxygen was maintained at 1–2 l/min. RFA was initiated after stable anesthesia was attained.

D-sorbitol was purchased from Baxter Ltd. (Tokyo, Japan). D-sorbitol RFA was performed under real-time ultrasonographic guidance with a 3.5 MHz sector probe (Power Vision 5000; Toshiba Medical, Tokyo, Japan) using the Cool-tip RF ablation system (Radionics, Burlington, MA, USA). For the D-sorbitol RFA treatment, a 16-gauge RFA needle was first inserted into the center of the tumor and a 18-gauge PEI needle was then introduced through the same hole of the attachment beside the echo probe. D-sorbitol 3% was slowly injected into the lesion through the side hole of the handpiece. The maximum total injected volume of D-sorbitol was 20 ml and RFA was terminated if the threshold of impedance was exceeded. RFA was performed for 20 W in 2 min, 30 W in 2 min and 50 W until the end. The maximum allowable output power was 50 W under the impedance control injecting D-sorbitol.

The energy required for ablation was calculated as follows: Energy (J) = Watt (W) × duration of ablation (sec). Animal livers containing RFA-induced coagulation were sliced in the transverse plane perpendicular to the electrode tracks. The approximation volume of the whole coagulated necrosis area in cm³ was calculated as follows:

Whole coagulation volume = 4/3π × r1 × r2 × r3,

where r1, longest diameter/2; r2, shortest diameter/2; and r3, height/2, all measured in cm. Two hepatologists with 10 years of clinical experience, who were blinded to the information regarding the RFA techniques used in the study, measured these parameters.

All data are presented as means ± standard deviation. Statistical significance was set at P<0.05 for Student's unpaired t-tests.

RFA and D-sorbitol RFA were performed in 5 different parts of pig livers. All RFA techniques were performed using the Cool-tip RF system and 20 ml of D-sorbitol was injected into the lesion through a 18-gauge PEI needle during RFA (Fig. 1A and B, right). The results of the comparison of the volume of coagulation necrosis and the total delivered energy for inducing coagulation between the two groups are summarized in Table I. The volumes of coagulation necrosis in the D-sorbitol RFA group (Fig. 1A and B, right) were significantly higher compared with those in the RFA group (Fig. 1A and B, left). Of note, the total delivered energy in the D-sorbitol RFA group was also higher compared with that in the RFA group (Table I).

The effect of D-sorbitol RFA was also examined in vivo using a dog model. The volumes of coagulation necrosis in the D-sorbitol RFA group were clearly higher compared with those in the RFA group in the dog model in vivo (Table II). In addition, the total delivered energy in the D-sorbitol RFA group was also higher compared with that in the RFA group (Table II). No significant complications, such as bleeding or damage, were observed during the procedures and 3 days after D-sorbitol RFA.