Lipo-prostaglandin E1 Improves Renal Hypoxia Evaluated by BOLD-MRI in Patients With Diabetic Nephropathy

R2* maps were calculated using the in-house software in Matlab (version 7.0, The MathWorks, Inc., Cambridge, MA, USA) on a pixel-by-pixel basis by fitting㏑(signal intensity) versus echo time. The R2* map was windowed from 0 to 60 s-1.

Regions of interest (ROIs) with unfixed size (7 to 53 pixels) were defined at the upper, middle and lower poles of both kidneys in the medulla and cortex in the renal hilum plane based on the anatomical images. Three ROIs were placed in the medulla and cortex, respectively, carefully avoiding vessels, renal sinus and susceptibility artifacts, yielding a total of 6 ROIs in both kidneys. The values in each ROI were averaged for the bilateral cortex and medulla. To prevent significant bias from measurement variability, each sample was measured independently by three professional radiologists with more than 5 years of experience who were blinded to the clinical results. Thereafter, assuming the tests were coincident, the results were the averaged.

Inclusion criteria

• Clinical diagnosis of chronic kidney disease

Exclusion criteria

• Intolerance to lipo-PGE1, contraindication for MRI, concurrent urinary tract infection, acute kidney injury, acute cardiovascular and cerebrovascular complications and diabetic emergencies, with the use of corticosteroid and diureses.