• Tetsuo Koyama, Masao Tsuji, Hiroyuki Nishimura, Hiroji Miyake, Takehisa Ohmura, and Kazuhisa Domen.
• Department of Rehabilitation Medicine, Nishinomiya Kyoritsu Neurosurgical Hospital, Hyogo, Japan. ytkoyama@bd6.so-net.ne.jp
• J Stroke Cerebrovasc Dis. 2013 Jan 1;22(1):72-9.

BackgroundMagnetic resonance-diffusion tensor imaging (DTI) was used to predict motor outcome for patients with intracerebral hemorrhage. We compared the predictive accuracy of data sampled from the cerebral peduncle with data from the corona radiata/internal capsule. This study included 32 subjects with thalamic or putaminal hemorrhage or both.MethodsDTI data were obtained on days 14 to 18. Mean values of fractional anisotropy (FA) within the cerebral peduncle and the corona radiata/internal capsule were analyzed using a computer-automated method. Applying ordinal logistic regression analyses, the ratios between FA values in the affected and unaffected hemisphere (rFA) were modeled in relation to motor outcome scores at 1 month after onset, assessed using the Medical Research Council (MRC) scale (0 = null to 5 = full).ResultsFor both cerebral peduncle and corona radiata/internal capsule, the relationships between rFA and MRC matched logistic probabilities. While cerebral peduncle rFA values had statistically significant relationships with MRC scores (upper extremity R(2) = 0.271; lower extremity R(2) = 0.191), rFA values for the corona radiata/internal capsule showed less significant relationships (upper extremity R(2) = 0.085; lower extremity R(2) = 0.080). When estimated cerebral peduncle rFA values were <0.7, estimated probability of MRC 0 to 2 was close to 85% for the upper and 60% for the lower extremities. Meanwhile, when estimated rFA values were >0.9, estimated probability for MRC 4 to 5 nearly equaled 50% for the upper and 60% for the lower extremities.ConclusionsFA values from within the cerebral peduncle more accurately predicted motor outcome and is a promising technique for clinical application.Copyright © 2013 National Stroke Association. Published by Elsevier Inc. All rights reserved.

34 keywords...

hide…