Percutaneous microwave ablation (MWA) is an effective and minimally invasive treatment for small tumors. To achieve local disease control, the entire tumor and a surrounding safety margin must be destroyed. Power and application time are chosen based on manufacturer-provided data, usually obtained from ex vivo animal models. However, ex vivo tissues differ from in vivo condition due to compositional changes and absence of heat dissipation by blood flow. This study aimed to compare in vivo ablation zone sizes in swine with those predicted by the device manufacturer.

Five pigs underwent 40 MWA procedures using various power-time-organ combinations; 18 hepatic and 20 renal zones were evaluable. All procedures were performed with devices from a single manufacturer (ECO Microwave System Co, Nanjing, China). After euthanasia, the ablation zones were excised and sliced. For each ablation, the slice showing the largest dimensions was selected to measure x and y diameters and used to compute the ablated surface area.

For 7 of 8 power-time-organ combinations, significant differences were found between predicted and measured surface areas (P<0.05), with deviations ranging from -45% to +54%. The overall mean absolute differences between measured and predicted ablation sizes in the x and y dimensions and the ablation surface area were 7.6 ± 4.6 mm (28% ± 19%), 5.8 ± 4.3 mm (18% ± 13%) and 273 ± 210 mm² (39% ± 34%), respectively.

Manufacturer-provided predictive data for MWA zone size may lack reliability. Intraoperative and postoperative monitoring of ablation zone size is crucial to ensure complete tumor destruction with adequate margins.