Comparison of Conventional and Cone-Beam CT for Monitoring and Assessing Pulmonary Microwave Ablation in a Porcine Model

Ece Meram; Colin Longhurst; Chris L Brace; Paul F Laeseke

doi:10.1016/j.jvir.2018.04.035

Comparison of Conventional and Cone-Beam CT for Monitoring and Assessing Pulmonary Microwave Ablation in a Porcine Model

J Vasc Interv Radiol. 2018 Oct;29(10):1447-1454. doi: 10.1016/j.jvir.2018.04.035. Epub 2018 Sep 11.

Authors

Ece Meram¹, Colin Longhurst², Chris L Brace³, Paul F Laeseke⁴

Affiliations

¹ Section of Interventional Radiology, Department of Radiology, University of Wisconsin, 600 Highland Avenue, D4-352, Madison, WI 53792. Electronic address: ecemeram@gmail.com.
² Department of Radiology, and Department of Biostatistics and Medical Informatics, University of Wisconsin, Madison, Wisconsin.
³ Tumor Ablation Laboratory, University of Wisconsin, Madison, Wisconsin.
⁴ Section of Interventional Radiology, University of Wisconsin, Madison, Wisconsin.

PMID: 30217749
DOI: 10.1016/j.jvir.2018.04.035

Abstract

Purpose: To compare cone-beam computed tomography (CT) with conventional CT for assessing the growth and postprocedural appearance of pulmonary microwave ablation zones.

Materials and methods: A total of 17 microwave ablations were performed in porcine lung in vivo by applying 65 W for 5 minutes through a single 17-gauge antenna. Either CT (n = 8) or CBCT (n = 9) was used for guidance and ablation zone monitoring at 1-minute intervals. Postprocedural noncontrast images were acquired with both modalities. Three independent readers measured the length, width, cross-sectional area, and circularity of the ablation zones on gross tissue samples and CT and cone-beam CT images. The measurements were compared via linear mixed-effects models for postprocedural appearance and with a polynomial mixed effects model for ablation zone growth curves.

Results: On postprocedural images, the differences between cone-beam CT and CT in mean length (3.84 vs 3.86 cm; Δ = -0.02; P = .70), width (2.61 vs 2.56 cm; Δ = 0.06; P = .46), area (7.84 vs 7.65 cm²; Δ = 0.19; P = .35), and circularity (0.85 vs 0.85; Δ = 0.01; P = .62) were not statistically significant after accounting for intersubject and interrater variability. Also, there was no significant difference between CT and cone-beam CT growth curves of the ablation zones during monitoring in terms of length (p_Int. = 1.00; p_Lin.Slope = 0.52; p_Quad.Slope = 0.69); width (p_Int. = 0.83; p_Lin.Slope = 0.98; p_Quad.Slope = 0.79), area (p_Int. = 0.47; p_Lin.Slope = 0.27; p_Quad.Slope = 0.57), or circularity (p_Int. = 0.54; p_Lin.Slope = 0.74; p_Quad.Slope = 0.80). Both CT and cone-beam CT overestimated gross pathologic observations of ablation length, width, and area (P < .001 for all).

Conclusions: Cone-beam CT was similar to conventional CT when assessing the growth, final size, and shape of pulmonary microwave ablation zones and may be useful for monitoring and evaluating microwave ablations in the lung.

Publication types

Comparative Study

MeSH terms

Ablation Techniques*
Animals
Cone-Beam Computed Tomography*
Female
Lung / diagnostic imaging
Lung / surgery*
Microwaves*
Models, Animal
Observer Variation
Predictive Value of Tests
Radiography, Interventional / methods*
Reproducibility of Results
Sus scrofa
Time Factors
Tomography, X-Ray Computed*