• ablation
• carbonization
• liver percutaneous treatment
• microwave ablation
• percutaneous ablative technique

Purpose: To retrospectively evaluate the relationship between microwave (MW) ablation operative technical parameters, volume of carbonization and volume of ablation.

Materials and methods: We retrospectively reviewed the radiological charts and follow-up imaging evaluations of all the patients who underwent MW ablations for primary and secondary liver lesions at our Institution in the period between November 2015 and April 2017.
All the ablations were performed with a 2.45GHz MW microwave generator (HS AMICA, HS Hospital Service, Rome, Italy) using cooled mini-choked antennas (14-gauge and 16-gauge).
The following technical parameters were recorded for all the procedures:
- power (P) released by the generator, expressed in watts (W);
- ablation time (T), defined as the the duration of ablation, expressed in seconds (S);
- cumulative energy deployed (total energy, TE), expressed in kilojoules (KJ), calculated using the formula: TE = (P*T).

At computed tomography (CT) imaging performed at least 4 weeks after the ablative procedure, the hypo-attenuating area after contrast medium administration was considered representative of coagulative necrosis, while the central inside zone of hyper-attenuation in unenhanced scans coincided with carbonization.
Visual conspicuity of carbonization was qualitatively assessed in all cases; patient population was then sub-grouped according to well-definition and ill-definition of the carbonization zone.
Measurements of both carbonized and thermocoagulated areas were obtained at first follow-up CT scan. In particular, maximal post-procedural diameters (long axis - AP, short axis - LL and craniocaudal extension) and volumes of both the areas were noted.
Correlation analysis were performed to compare diameters and volumes of both ablation and carbonization zones with time of ablation, power employed and total energy delivered.

Results: A moderately significant correlation between TE and the resulting volume of necrosis was found (ρ=0.52).
Bivariate analysis showed a strong linear relationship between necrosis volume and carbonization volume; this finding was also confirmed by Spearman rank correlation (ρ=0.57).
The individual carbonization diameters considered as independent variables correlated with the equivalent necrosis diameters, in particular in AP and LL orientation (ρ=0.64 and 0.63, respectively).
A significant linear relationship between carbonization volume and reference ablation parameters was found; such relationship is preserved even in the subgroup of ill-defined carbonization area.
However, Spearman rank correlation did not show any correlation between carbonization diameters, carbonization volume and reference ablation parameters.

Conclusions: Carbonization represents a peculiar feature of MW ablation attributable to the higher temperatures reached in the treated lesion respect to other ablative techniques. Our results showed a strong correlation between ablation and carbonization diameters and volumes, which might suggest that carbonization could represent an indirect marker of the overall tumoral tissue ablated by MW. Therefore, the identification of carbonization at imaging may represent a way to monitor the ablative procedure and to obtain a qualitative feedback of the treatment efficacy.