BOLD quantified renal pO2 is sensitive to pharmacological challenges in rats.

Purpose: Blood oxygen level-dependent (BOLD) MRI has been effectively used to monitor changes in renal oxygenation. However, R2* (or T2*) is not specific to blood oxygenation and is dependent on other factors. This study investigates the use of a statistical model that takes these factors into account and maps BOLD MRI measurements to blood pO2.

Methods: Spin echo and gradient echo images were obtained in six Sprague-Dawley rats and R2 and R2* maps were computed. Measurements were made at baseline, post-nitric oxide synthase inhibitor (L-NAME), and post-furosemide administration. A simulation of each region was performed to map R2' (computed as R2*-R2) to blood pO2.

Results: At baseline, blood pO2 in the outer medulla was 30.5 ± 1.2 mmHg and 51.9 ± 5.2 mmHg in the cortex, in agreement with previous invasive studies. Blood pO2 was found to decrease within the outer medulla following L-NAME (P < 0.05) and increase after furosemide (P < 0.05). Blood pO2 in the cortex increased following furosemide (P < 0.05).

Conclusions: Model-derived blood pO2 is sensitive to pharmacological challenges, and baseline pO2 is comparable to literature values. Reporting pO2 instead of R2* could lead to a greater clinical impact of renal BOLD MRI and facilitate the identification of hypoxic regions. Magn Reson Med 78:297-302, 2017. © 2016 International Society for Magnetic Resonance in Medicine.