An absolute measurement of brain water content using magnetic resonance imaging in two focal cerebral ischemic rat models.

Abstract

Magnetic resonance imaging (MRI) was utilized to obtain absolute estimates of regional brain water content (W), and results were compared with those obtained with conventional wet/dry measurements. In total, 31 male Long-Evans rats were studied and divided into two groups based on the surgical procedures used to induce cerebral focal ischemia: suture (n = 18) and three-vessel ligation (TVL: n = 13) groups. Both relative spin density and T1 were extracted from the acquired MR images. After correcting for radiofrequency field inhomogeneities, T2* signal decay, and temperature effects, in vivo regional brain water content, in absolute terms, was obtained by normalizing the measured relative brain spin density of animals to that of a water phantom. A highly linear relationship between MR-estimated brain water content based on the normalized spin density and wet/dry measurements was obtained with slopes of 0.989 and 0.986 for the suture (r = 0.79) and TVL (r = 0.83) groups, respectively. Except for the normal subcortex of the TVL group (P < 0.02) and the normal hemisphere of the suture group (P < 0.003), no significant differences were observed between MR-estimated and wet/dry measurements of brain water content. In addition, a highly linear relationship between MR-measured R1 (= 1/T1) and 1/W of wet/dry measurements was obtained. However, slopes of the linear regression lines in the two groups were significantly different (P < 0.02), indicating that different R1 values were associated with the same water content depending on the model. These results show that an absolute measurement of in vivo regional brain water content can be obtained with MRI and potentially serves as a noninvasive means to monitor different therapeutic interventions for the management of brain edema subsequent to stroke and head trauma.