Oral Abstracts
13th Annual Global Embolization Symposium & Technologies
Damian Hall
University of Minnesota
Purpose : The efficacy of cryoablation in the treatment of renal cell carcinoma (RCC) has been established. While the manufacturer has provided specifications for predicted ablation zone size in 38-degree gel, little data is available in how accurately these predictions translate to human procedures. The goal of this single center retrospective review is to evaluate the accuracy with which manufacturer predictions apply to in human use.
Material and Methods : All patients who underwent cryoablation between 2007 and 2017 were retrospectively reviewed. Of the 79 RCC patients 33 patients had 2, 3, or 4 of the same cryoablation probe placed in a configuration consistent with manufacturer provided data. This included 28 patients who had IceRod™ CX (BTG, London, UK) needles placed and 5 with IceSphere™ CX (BTG, London, UK) needles placed. The patients charts were retrospectively reviewed to ascertain baseline laboratory values, imaging characteristics, and procedural data. The ice ball formation was measured in 2 directions anterior posterior (AP) (parallel to the ablation shaft) and right-left (RL) (perpendicular to the ablation shaft) on procedural axial computed tomography exams. Percentage of predicted was calculated by dividing the realized ice ball size by the manufacturer prediction of 0 degrees.
Results : The percentage of predicted value in the AP direction was 67.7% ± 11.8%, while it was 59% ± 10.9% in the RL direction. When only evaluating IceRod™ CX patients the percentage of predicted value in the AP direction 68.2% ± 11.7%, while it was 57.6% ± 10.6% in the RL direction. When only evaluating the 5 patients with IceSphere probes placed the percentage of predicted in the AP direction was 60% ± 7.8%, while it was 70.6% ± 9.0% in the RL direction.
Conclusions : While the manufacturer predicted values of cryoablation procedures are fairly accurate, they tend to over predict the size of the 0 degree iceball.