Accuracy of liver ablation zone prediction in a single 2450 MHz 100 Watt generator model microwave ablation system: An in human study

Purpose: To compare manufacturer provided predictions and realized ablation dimensions in the liver using one 2450 MHz 100 Watt generator model microwave ablation (MWA) system. Materials and method: Between 1/1/2015 and 2/1/2018, MWAs were performed in 86 patients who underwent a total of 103 MWAs with a single MWA system. There were 64 men and 22 women with a mean age of 63.9 ± 9.9 (SD) years (range: 30–88 years). Demographic, procedural, and outcomes data was recorded. The manufacturer predicted ablation zone sizes in three dimensions (anterior-posterior [AP], transverse [TR], and cranial caudal [CC]) were recorded and then compared to the actual ablation zone sizes at one month follow-up imaging. Results: MWAs were most commonly performed to treat hepatocellular carcinoma (92/103, 89.3%). Dividing the actual ablation size by the manufacturer prediction in the AP, TR, and CC directions resulted in a mean of 88.3 ± 20.6 (SD) % (range: 33.3–156.4%), 80.2 ± 26.5 (SD) % (range: 29.6–182.9%), and 86.7 ± 25.1 (SD) % (range: 37–186.1%), respectively. The realized AP direction was statistically closer to the manufacturer prediction than the TR (P < 0.01). Ablation Watt setting of 100 Watts resulted in more accurate predictions than the 75 or 45 Watt settings in the AP direction (P = 0.03). Conclusions: This 2450 MHz 100 Watt generator MWA system manufacturer provided model fairly accurately predicts ablation zone dimensions, but tends to over predict realized dimensions in this mainly hepatocellular carcinoma, and therefore cirrhotic, cohort. The TR is the most inaccurately predicted dimension and manufacturer predictions appear to be best in the 100 W setting, important aspects for interventionalists to consider during ablation planning and execution.