Effects of perfusion on radiofrequency ablation in swine kidneys

Abstract

Purpose: To evaluate the effect of vascular occlusion on the size of radiofrequency (RF) ablation lesions and to evaluate embolization as an occlusion method.

Materials and methods: The kidneys of six swine were surgically exposed. Fifteen RF ablation lesions were created in nine kidneys by using a 2-cm-tip single-needle ablation probe in varying conditions: Seven lesions were created with normal blood flow and eight were created with blood flow obstructed by means of vascular clamping (n = 5) or renal artery embolization (n = 3). The temperature, applied voltage, current, and impedance were recorded during RF ablation. Tissue-cooling curves acquired for 2 minutes immediately after the ablation were compared by using regression analysis. Lesions were bisected, and their maximum diameters were measured and compared by using analysis of variance.

Results: The mean diameter of ablation lesions created when blood flow was obstructed was 60% greater than that of lesions created when blood flow was normal (1.38 cm +/- 0.05 [standard error of mean] vs 0.86 cm +/- 0.07, P <.001). The two methods of flow obstruction yielded lesions of similar mean sizes: 1.40 cm +/- 0.06 with vascular clamping and 1.33 cm +/- 0.07 with embolization. The temperature at the probe tip when lesions were ablated with normal blood flow decreased more rapidly than did the temperature when lesions were ablated after flow obstruction (P <.001), but no significant differences in tissue-cooling curves between the two flow obstruction methods were observed.

Conclusion: Obstruction of renal blood flow before and during RF ablation resulted in larger thermal lesions with potentially less variation in size compared with the lesions created with normal nonobstructed blood flow. Selective arterial embolization of the kidney vessels may be a useful adjunct to RF ablation of kidney tumors.

Figure 1

Ablation lesions (arrows) created with renal vasculature clamping (left) and normal blood flow (right) in one swine by using a steady-state temperature of 98°C for 10 minutes. The lesion specimens were incubated for 10 minutes with 1% 2,3,5-triphenyltetrazolium chloride (red) solution, which stains viable tissue red, and then fixed in formalin. The dimensions (maximum width × depth) of the lesions were 1.5 × 2.0 cm with vasculature clamping and 1.0 × 2.1 cm with normal blood flow.

Figure 2

Ablation lesions (arrows) created (a) following selective embolization of the renal vasculature to angiographic stasis and (b) with normal blood flow in one swine by using a steady-state temperature of 98°C for 10 minutes. The lesion specimens shown were fixed in formalin and not stained. The dimensions (maximum width × depth) of the lesions were 1.4 × 2.3 cm following embolization and 0.6 × 2.3 cm with normal blood flow.

Figure 3

Tissue-cooling curves with normal flow, vascular clamping, and embolization during the first 110 seconds after the completion of RF energy deposition. Symbols represent the mean values for each treatment at each time point. Error bars represent standard errors of the mean for each treatment at each time point. The lines represent the two-phase exponential decay curves for each treatment, calculated by using nonlinear regression. The tissue-cooling curves for the two blood flow obstruction methods were similar. There was more rapid cooling of the lesions created with normal blood flow; the tissue-cooling curve for this treatment rapidly diverged from the curves for blood flow obstruction.