Search

The purpose of this study was to compare the accuracy (in terms of ultrasound-guided probe placement) and the effectiveness (in terms of pathologic tumor-free margin) of laparoscopic vs open radiofrequency (RF) ablation.Using a previously validated tissue-mimic model, 1-cm simulated hepatic tumors were ablated in 10 pigs randomized to open or laparoscopic techniques. Energy was applied until tissue temperature reached 100 degrees C (warm-up) and thereafter for 8 min. A pathologist blinded to technique examined all specimens immediately after treatment. Analysis was by Fisher's exact test and the Mann-Whitney U test; p0.05 was considered significant.Off-center distance (3.5 +/- 1.6 vs 4.2 +/- 1.4 mm), size (24.7 +/- 3.1 vs 25.6 +/- 3.8 mm), symmetry (40% vs 73%), margin positivity (33% vs 9%), and margin distance (1.1 +/- 1.2 vs 2.2 +/- 1.6 mm) were not significantly different between laparoscopic (n = 15) and open (n = 11) ablations, respectively. The proportion of round/ovoid lesions (20% vs 64%) was lower (p = 0.043), and warm-up time (20.2 +/- 14.0 vs 10.7 +/- 7.5) was longer (p = 0.049) for the laparoscopic than for the open groups, respectively.Accurate probe placement can be achieved using laparoscopic and open RF ablation techniques. The physiologic effects of laparoscopy may alter ablation shape and warm-up time. Additional studies are needed to establish effective ways of achieving complete tumor destruction.

Radiofrequency ablation requires accurate probe placement using ultrasound guidance. The purpose of this study was to develop an in vivo tumor-mimic model for learning open and laparoscopic radiofrequency ablation. Tumor-mimics were created in ex vivo porcine livers by injecting a mixture of 3% agarose, 3% cellulose, 7% glycerol, and 0.05% methylene blue, which formed 1 cm hyperechoic, discrete lesions on ultrasound. Open and laparoscopic (using a box-trainer) ablation techniques were practiced. In vivo experiments were then conducted in 10 pigs. Three tumor-mimics were created in each animal using a laparoscopic approach. Lesions were characterized sonographically, ablated using an open (n = 5) or laparoscopic (n = 5) approach, and examined pathologically. An ablation in normal liver tissue was performed as a control. Tissue impedance was recorded. Target creation took 81 minutes per animal and 96% of injections were successful. Tissue impedance (48.8 +/- 5.8 vs. 49.6 +/- 5.4) and ablation size (25.1 +/- 3.4 vs. 24.3 +/- 5.1) were not significantly different for controls (n = 8) and tumor-mimics (n = 26), respectively. One animal died of a pulmonary embolism following injection of agarose into a hepatic vein. The agarose-based tissue-mimic creates realistic sonographic targets for learning ultrasound-guided open and laparoscopic radiofrequency ablation in an in vivo model.