The spatial heterogeneity of renal perfusion data was analyzed with arterial spin labeling (ASL) data sets in a cohort of subjects with moderately impaired kidney function (ie, glomerular filtration rate >30 mL/min/1.73 m2) versus a cohort of healthy volunteers. The potential diagnostic value of a detailed histogram analysis of such perfusion data for detection of mild renal dysfunction was investigated.
Materials and Methods
Eight healthy volunteers and 9 patients with mild renal dysfunction (chronic kidney disease stages 1-3) were included in the study. All subjects underwent ASL perfusion measurements with a 1.5-T magnetic resonance scanner using a flow-sensitive alternating inversion recovery labeling scheme with true fast imaging in steady-state precession data readout. Quantitative perfusion maps were generated using extended Bloch equations. Histogram analysis was performed to quantify the metrics of the perfusion of the renal cortex and the entire parenchyma, respectively. Mean perfusion value (μ), SD of the mean value (σ), peak height (PH), peak position (PP), skewness (s), and kurtosis (k) were computed to describe the distribution of the perfusion values.
A significant difference was found in the mean perfusion values computed for the cortex and the parenchyma between healthy volunteers (cortex, 329 ± 53 mL/100 g/min; parenchyma, 301 ± 51 mL/100 g/min) and patients (cortex, 263 ± 81 mL/100 g/min; parenchyma, 244 ± 77 mL/100 g/min). The histogram analysis of the cortical perfusion values also showed a significant difference (P < 0.05) in the main histogram measures between healthy volunteers (PP = 368 ± 65 mL/100 g/min; s = −0.543 ± 0.298; k = 0.371 ± 0.590) and patients (PP = 237 ± 115 mL/100 g/min; s = −0.125 ± 0.581; k = −0.151 ± 0.561).
Moderate renal dysfunction is associated with a significant change in the distribution of cortical perfusion values and a reduction of blood perfusion for both the parenchyma and the cortex. The preliminary results reported in this study suggest the importance of a regional assessment of renal perfusion. Histogram analysis of ASL data may help to detect chronic kidney disorders and to monitor their progression in a clinical setting.