Assessment of Blood Oxygen Level-Dependent Magnetic Resonance Imaging for Differentiating Renal Dysfunction From...

The aim of the study was to investigate the potential of T2* histogram (HIST) in the analysis of blood oxygen level-dependent magnetic resonance imaging for differentiating a renal dysfunction group from a control group.

This study consisted of 13 control subjects and 20 patients with renal dysfunction. T2* maps were analyzed using both HIST and the conventional method. For the HIST, each region of interest covering renal parenchyma was applied to T2* map to generate a T2* HIST. The T2* HISTs were classified into type 1, 2, 3, and 4 on the basis of their morphologies. In each T2* HIST, the parameters were acquired from the peak with the smallest mean T2* values, which acted as the medullary T2* values, and from the peak with largest mean T2* values, which acted as the cortical T2* values. For the conventional method, small regions of interest were placed in the cortex and medulla for the quantitative analysis of the cortical and medullary T2* values, respectively.

The type distribution of the T2* HISTs was significantly different between the 2 groups (control group: type 1, 0; 2, 4; 3, 5; and 4, 17 vs renal dysfunction group: type 1, 4; 2, 1; 3, 33; and 4, 2; P < 0.05). The medullary T2* values measured using both methods were significantly higher in the renal dysfunction group than that in the control group (29.38 [6.44] vs 22.79 [4.22] milliseconds for HIST, 30.61 [8.65] vs 21.37 [4.88] milliseconds for conventional method; P < 0.05). The area under the receiver operating characteristic curve for the medullary T2* measured using HIST was not greater than that using the conventional method (0.89 vs 0.82, P > 0.05).

Histogram can automatically characterize the T2* map of renal parenchyma, thereby reducing the tediousness of placing ROIs and making it comparable with the conventional method in its ability to distinguish the renal dysfunction group from the control group.