Morphometric characterization and temporal temperature measurements during hepatic microwave ablation in swine

PURPOSE: Heat-induced destruction of cancer cells via microwave ablation (MWA) is emerging as a viable treatment of primary and metastatic liver cancer. Prediction of the impacted zone where cell death occurs, especially in the presence of vasculature, is challenging but may be achieved via biophysical modeling. To advance and characterize thermal MWA for focal cancer treatment, an in vivo method and experimental dataset were created for assessment of biophysical models designed to dynamically predict ablation zone parameters, given the delivery device, power, location, and proximity to vessels. MATERIALS AND METHODS: MWA zone size, shape, and temperature were characterized and monitored in the absence of perfusion in ex vivo liver and a tissue-mimicking thermochromic phantom (TMTCP) at two power settings. Temperature was monitored over time using implanted thermocouples with their locations defined by CT. TMTCPs were used to identify the location of the ablation zone relative to the probe. In 6 swine, contrast-enhanced CTs were additionally acquired to visualize vasculature and absence of perfusion along with corresponding post-mortem gross pathology. RESULTS: Bench studies demonstrated average ablation zone sizes of 4.13±1.56cm2 and 8.51±3.92cm2, solidity of 0.96±0.06 and 0.99±0.01, ablations centered 3.75cm and 3.5cm proximal to the probe tip, and temperatures of 50 ºC at 14.5±13.4s and 2.5±2.1s for 40W and 90W ablations, respectively. In vivo imaging showed average volumes of 9.8±4.8cm3 and 33.2±28.4cm3 and 3D solidity of 0.87±0.02 and 0.75±0.15, and gross pathology showed a hemorrhagic halo area of 3.1±1.2cm2 and 9.1±3.0cm2 for 40W and 90W ablations, respectfully. Temperatures reached 50ºC at 19.5±9.2s and 13.0±8.3s for 40W and 90W ablations, respectively. CONCLUSION: MWA results are challenging to predict and are more variable than manufacturer-provided and bench predictions due to vascular stasis, heat-induced tissue changes, and probe operating conditions. Accurate prediction of MWA zones and temperature in vivo requires comprehensive thermal validation sets.