The majority of chronic kidney diseases are characterized by renal fibrosis¹. While biopsy is the gold standard for fibrosis assessment, it is unsuitable for early detection of fibrotic renal disease, particularly in humans. T2-weighted MRI is often used to detect tissue damage², but it is not a direct assessment of tissue function. In this study, we examined renal perfusion in healthy and fibrotic rats using arterial spin labeling (ASL) at multiple inversion times (TIs) to directly estimate fibrotic damage to kidney function and compared it to renal T2 as well as urinary and histological fibrosis markers.

Following IACUC approval, fibrosis was induced in the renal cortex of 11 male Sprague Dawley rats (Charles River Laboratories) through injection of 1.75 mg/kg monoclonal antibody OX-7. Two other rats did not receive the injection and were used as healthy controls. Renal T2-weighted and ASL MRI, a 24-hour urine collection protocol to measure proteinuria, and renal biopsy to assess the degree of cortical fibrosis was performed 5 days after OX-7 injection for 6 of the fibrotic rats and 12 days after OX-7 injection for the other 5 fibrotic rats.

Rats were sedated with isoflurane (Fluriso; Vet One) during the MR examination. MR imaging was performed using a wrist coil on a 3T clinical scanner (TimTrio; Siemens). Sets of eight T2-weighted images were acquired from axial slices through each kidney using a turbo spin-echo sequence: TR 400ms, TE 5.8, 29, 52, 75, 98, 122, 145, 168ms, flip angle 180°, matrix 256x256, resolution 0.78x0.78mm. Each series of T2-weighted images was fit pixel-wise to the T2 decay function to generate a T2 map. Cortical T2 was determined from the T2 map by averaging values within an ROI defined over the renal cortex. ASL images were acquired from the same slices as the T2 images using a flow-sensitive alternating inversion recovery (FAIR) protocol: TR 4.7ms, TE 2.35ms, TIs 300, 500, 800, 1000, 1300, 1500ms, flip angle 180°, matrix 256x256, resolution 0.78x0.78mm, bSSFP readout, and 6s inter-image time. For the tag and control images at each TI, the signal intensity was averaged within the same cortical ROIs as for T2 analysis. Subtraction of the control from the tag signal-vs-TI curve generated the ASL difference signal. Using a tracer-kinetic model of the multi-TI ASL difference signal, similar to that proposed by Buxton et al³, renal plasma flow (RPF) was measured in the cortex by calculating the slope of the difference signal normalized by T1 relaxation.

After MR examination, rats were housed in metabolic cages for 24 hours to allow for urine collection. Urinary albumin and general protein concentrations were measured using a DC2000+ reagent kit (Bayer) and the Bradford method (Bio-Rad), respectively. Rats were then sacrificed and their kidneys removed and fixed in formalin. Sections of cortical tissue were embedded in paraffin and stained with periodic acid-Schiff (PAS) solution to visualize the glomeruli. Subsections of these samples containing 20 glomeruli were photographed under 400x magnification. Using semi-automatic color-image analysis software (Image J; NIH), fibrosis was quantified as the PAS-positive mesangial area over the total area of the glomerular tuft. RPF, T2, urinary albumin and protein excretion, and histological fibrosis measurements were compared between control rats and those 5-days and 12-days after fibrosis induction. Two-sided t-tests (α=0.05) were used to test for significant differences between groups.

Compared to the controls, significantly elevated levels of albumin and general protein were measured in the urine of fibrotic rats (Figure 1). Photomicrographs of glomeruli from healthy and fibrotic rats are shown in Figures 2A-2C, demonstrating increased PAS-positive mesangial matrix in the diseased groups. The average degree of fibrosis (percentage of glomerular tuft) is quantified in Figure 2D, showing significantly greater fibrosis in rats treated with monoclonal antibody OX-7. ASL images from healthy and diseased rats are compared in Figures 3A and 3B, illustrating a significant drop in average cortical perfusion compared to controls (Figure 3C). Cortical T2 values were also significantly greater in fibrotic rats than in controls (Figure 4). No significant difference was observed between the 5-day and 12-day fibrosis groups with any metric. These results are summarized in Table 1.Cortical RPF and T2 changes both revealed fibrotic damage to the kidneys, in parallel with observations from urinary and histological fibrosis markers. Unlike T2, however, RPF is a direct measurement of renal function. ASL may therefore provide improved assessment of fibrotic kidney damage, and in conjunction with structural imaging techniques like T2-weighted MRI, prove useful for the early detection of renal disease.