In vivo PO2 imaging in the porcine model with perfluorocarbon F-19 NMR at low field

Quantitative pO2 imaging in vivo has been evaluated utilizing F-19 NMR in the porcine model at 0.14 T for the lungs, liver, and spleen following i.p. administration of the commercial perfluorotributylamine (FC-43)-based perfluorocarbon (PFC) emulsion, Oxypherol-ET. Calculated T1 maps obtained from a two spin-echo saturation recovery/inversion recovery (SR/IR) pulse protocol are converted into quantitative pO2 images through a temperature-dependent calibration curve relating longitudinal relaxation rate (1/T1) to pO2. The uncertainty in pO2 for a T1 measurement error of +/- 5% as encountered in establishing the calibration curves ranges from +/- 10 torr (+/- 40%) at 25 torr to +/- 16 torr (+/- 11%) at 150 torr for FC-43 (37 degrees C). However, additional uncertainties in T1 dependent upon the signal-to-noise ratio may be introduced through the SR/IR calculated T1 pulse protocol, which might severely degrade the pO2 accuracy. Correlation of the organ image calculated pO2 with directly measured pO2 in airway or blood pools in six pigs indicate that the PFC resident in lung is in near equilibrium with arterialized blood and not with airway pO2, suggesting a location distal to the alveolar epithelium. For the liver, the strongest correlation implying equilibrium was evident for venous blood (hepatic vein). For the spleen, arterial blood pO2 (aorta) was an unreliable predictor of pO2 for PFC resident in splenic tissue. The results have demonstrated the utility and defined the limiting aspects quantitative pO2 imaging in vivo using F-19 MRI of sequestered PFC materials.