Multimodal visibility (radiography, computed tomography, and magnetic resonance imaging) of microspheres for transarterial embolization tested in porcine kidneys.

ABSTRACT

OBJECTIVE: The objective of this study was to test multimodal visibility (radiography, computed tomography [CT], and magnetic resonance imaging [MRI]) of microspheres for transarterial embolization in porcine kidneys. MATERIALS AND METHODS: Currently available embolization particles (microspheres) were modified. A dense x-ray material (barium sulfate) was added to create visibility for radiography and CT. A magnetic substance (iron oxide) was additionally added to create visibility for MRI. This chemical modification was performed for particles with sizes of 100 ± 25 and 700 ± 50 µm. Three different prototypes per size class (samples A, B, and C) resulted, each with a different degree of barium sulfate but with the same degree of iron oxide. The currently available embolization particles with sizes of 100 ± 25 and 700 ± 50 µm were used as controls (sample control). Eight renal arteries of 4 pigs were embolized. Study end points were size distribution evaluated in vitro as well as qualitative and quantitative particle visibility evaluated in vivo. RESULTS: The size distribution of the particles with a size of 100 ± 25 µm was between 96 ± 11 µm for sample A and 102 ± 13 µm for the sample control without significant differences (n.s.). The size distribution of the particles with a size of 700 ± 50 µm was between 691 ± 20 µm for sample A and 716 ± 34 µm for sample C without significant differences (n.s.). For radiography, the particles with sizes of 100 ± 25 and 700 ± 50 µm for samples A, B, and C were definitely visible during the embolization. The sample control was definitely not visible. For CT and MRI (T1-weighted [T1w] and T2-weighted [T2w]), the particles with sizes of 100 ± 25 and 700 ± 50 µm for samples A, B, and C were definitely visible after the embolization. The sample control was definitely not visible. For CT, the signal-to-noise ratio for samples A, B, and C increased significantly after the embolization (eg, sample A, particles with a size of 100 ± 25 µm 66.5% ± 23.7%, P < 0.05). The signal-to-noise ratio for the sample control did not change after the embolization (eg, sample control, particles with a size of 700 ± 25 µm -0.2% ± 15.2%, n.s.). For MRI (T1w and T2w), the signal-to-noise ratio for samples A, B, and C decreased significantly after the embolization (eg, sample B, particles with a size of 700 ± 50 µm, T1w -72.9% ± 6.6%; P < 0.05). The signal-to-noise ratio for the sample control did not change after the embolization (eg, sample control, particles with a size of 100 ± 25 µm, T2w 6.2% ± 16.1%, n.s.). CONCLUSIONS: In this study, the chemical modification of the currently available microspheres for transarterial embolization resulted in a size distribution comparable with the currently available microspheres and created multimodal visibility for radiography, CT, and MRI.