Intraoperative ultrasound and tissue elastography measurements do not predict the size of hepatic microwave ablations

Image-guided thermal ablation is used to treat primary and secondary liver cancers. Evaluating completeness of ablation is difficult with standard intraoperative B-mode ultrasound. This study evaluates the ability of B-mode ultrasound (US) and tissue elastography to adequately measure the extent of ablation compared to pathologic assessment.

An in vivo porcine model was used to compare B-mode ultrasonography and elastography to pathologic assessment of the microwave ablation zone area. In parallel, intraoperative ablations in patients were used to assess the ability of B-mode US and elastographic measures of tissue strain immediately after ablation to predict ablation size, compared to postprocedural computed tomography (CT).

In the animal model, ablation zones appeared to decrease in size when monitored with ultrasound over a 10-minute span with both B-mode US and elastography. Both techniques estimated smaller zones than gross pathology, however, the differences did not reach statistical significance. Biopsies from the edges of the ablation zone, as assessed by US, contained viable tissue in 75% of the cases. In the human model, B-mode US and elastography estimated similar ablation sizes; however, they underestimate the final size of the ablation defect as measured on postprocedure CT scan (median area [interquartile range]: CT, 7.3 cm(2) [5.2-9.5] vs. US 3.6 cm(2) [1.7-6.3] and elastography 4.1 cm(2) [1.4-5.1]; P = .005).

Ultrasound and elastography provide an accurate gross estimation of ablation zone size but are unable to predict the degree of cellular injury and significantly underestimate the ultimate size of the ablation.