Is dual-phase C-arm CBCT sufficiently accurate for the diagnosis of colorectal cancer liver metastasis during liver intra-arterial treatment?

Utility of 'dual phase' cone beam computed tomography during radioembolisation in patients with hepatocellular carcinoma: what is really changing in flow dynamics before and after 90Y delivery?

Purpose

The aims of the study were: 1) to compare two phases of dual-phase cone beam computed tomography (DP-CBCT) achieved before and after Yttrium-90 (⁹⁰Y) administration and to evaluate additional benefits during radioembolisation (RE) procedures; and 2) to compare DP-CBCT with pre-procedure contrast enhanced cross-sectional images in terms of tumour detection.

Material and methods

Twenty-three hepatocellular carcinoma patients undergoing RE treatment were scanned with DP-CBCT consisting of early arterial (EA) and late arterial (LA) phases before and after ⁹⁰Y administration. The CT-like datasets were compared according to embolisation effect, enhancement patterns, lesion detectability, image quality, and artifacts by two interventional radiologists blinded to each other. The compatibility of the two radiologists was evaluated with kappa statistical analysis, and the difference between EA and LA phases was evaluated with marginal homogeneity test. Also, DP-CBCT images were compared with preprocedural cross-sectional images (CT/MRI).

Results

For 23 patients 92 data were acquired. Thirteen patients showed a decrease on post-embolisation images both visually and on Hounsfield unit (HU) measurements. No statistical difference was found for tumour detection between EA and LA phases (p = 1.0). Tumour enhancement was visually superior at LA phases whereas EA phases were better for arterial mapping for selective catheterisation. DP-CBCT images were not inferior to preprocedural cross-sectional imaging findings.

Conclusions

DP-CBCT is a promising tool for predicting tumour response to therapy and is not inferior to preprocedural cross-sectional imaging in terms of tumour detection. It allows better assessment during RE procedures because early phases provide good mapping for superselective catheterisation whereas late phases are better for visualisation of tumour enhancement.