OBJECTIVE: X-ray phase contrast imaging generates contrast from refraction of X-rays, enhancing soft tissue contrast compared to conventional absorption-based imaging. Our goal is to develop a contrast agent for X-ray in-line phase contrast imaging (PCI) based on ultrasound microbubbles (MBs), by assessing size, shell material, and concentration. METHODS: Polydisperse perfluorobutane-core lipid-shelled MBs were synthesized and size separated into five groups between 1 and 10 μm. We generated two size populations of polyvinyl-alcohol (PVA)-MBs, 2-3 μm and 3-4 μm, whose shells were either coated or integrated with iron oxide nanoparticles (SPIONs). Microbubbles were then embedded in agar at three concentrations: 5 × 107, 5 × 106 and 5 × 105 MBs/ml. In-line phase contrast imaging was performed at the Canadian Light Source with filtered white beam micro-computed tomography. Phase contrast intensity was measured by both counting detectable MBs, and comparing mean pixel values (MPV) in minimum and maximum intensity projections of the overall samples. RESULTS: Individual lipid-MBs 6-10 μm, lipid-MBs 4-6 μm and PVA-MBs coated with SPIONs were detectable at each concentration. At the highest concentration, lipid-MBs 6-10 μm and 4-6 μm showed an overall increase in positive contrast, whereas at a moderate concentration, only lipid-MBs 6-10 μm displayed an increase. Negative contrast was also observed from two largest lipid-MBs at high concentration. CONCLUSION: These data indicate that lipid-MBs larger than 4 μm are candidates for PCI, and 5 × 106 MBs/ml may be the lowest concentration suitable for generating visible phase contrast in vivo. SIGNIFICANCE: Identifying a suitable MB for PCI may facilitate future clinical translation.
OBJECTIVE: X-ray phase contrast imaging generates contrast from refraction of X-rays, enhancing soft tissue contrast compared to conventional absorption-based imaging. Our goal is to develop a contrast agent for X-ray in-line phase contrast imaging (PCI) based on ultrasound microbubbles (MBs), by assessing size, shell material, and concentration. METHODS:Polydisperse perfluorobutane-core lipid-shelled MBs were synthesized and size separated into five groups between 1 and 10 μm. We generated two size populations of polyvinyl-alcohol (PVA)-MBs, 2-3 μm and 3-4 μm, whose shells were either coated or integrated with iron oxide nanoparticles (SPIONs). Microbubbles were then embedded in agar at three concentrations: 5 × 107, 5 × 106 and 5 × 105 MBs/ml. In-line phase contrast imaging was performed at the Canadian Light Source with filtered white beam micro-computed tomography. Phase contrast intensity was measured by both counting detectable MBs, and comparing mean pixel values (MPV) in minimum and maximum intensity projections of the overall samples. RESULTS: Individual lipid-MBs 6-10 μm, lipid-MBs 4-6 μm and PVA-MBs coated with SPIONs were detectable at each concentration. At the highest concentration, lipid-MBs 6-10 μm and 4-6 μm showed an overall increase in positive contrast, whereas at a moderate concentration, only lipid-MBs 6-10 μm displayed an increase. Negative contrast was also observed from two largest lipid-MBs at high concentration. CONCLUSION: These data indicate that lipid-MBs larger than 4 μm are candidates for PCI, and 5 × 106 MBs/ml may be the lowest concentration suitable for generating visible phase contrast in vivo. SIGNIFICANCE: Identifying a suitable MB for PCI may facilitate future clinical translation.