| 제목 |
Prediction of Functional Motor Recovery using Diffusion Tensor Imaging in Subacute Brain Injury |
| 소속 |
Jeonbuk National University Medical School, Department of Physical Medicine & Rehabilitation1, Biomedical Research Institute of Jeonbuk National University Hospital, Research Institute of Clinical Medicine of Jeonbuk National University 2, Jeonbuk National University Hospital, Translational Research & Clinical Trial Center for Medical Device3 |
| 저자 |
Choong-Hee Roh 1,2*, Da-Sol Kim 1,2, Yu-Hui Won1,2, Sung-Hee Park1,2, Myoung-Hwan Ko1,2, Jeong-Hwan Seo1,2, Gi-Wook Kim1,2† |
Background
Prediction of functional level via diffusion tensor imaging (DTI) has been reported, but the evidence remains insufficient. This study aims to determine whether fiber number (FN) and fractional anisotropy (FA) measurements of DTI are sufficiently objective to assess motor recovery of the extremities using receiver operating characteristic (ROC) curves of motor evoked potential (MEP) and functional assessment [Medical Research Council (MRC), hand function test (HFTs), Modified Barthel Index (MBI)] and correlate this data.
Methods
We reviewed the charts of 70 unilateral hemiplegia patients with subacute brain injury from May 2018 to March 2020, who underwent DTI. For other evaluations, we reviewed the amplitude of the first dorsal interosseous (FDI) and tibialis anterior (TA) muscles in a MEP study, and functional assessment was reviewed. In DTI, a seed region of interest (ROI) was drawn at the corticospinal tract (CST) portion of the anterior mid-pons using the DTI studio software v.1.02. To evaluate CST, FN and FA of the affected and non-affected sides were measured. Differences between MEP amplitude and functional assessment of the affected and non-affected sides and FN and FA of DTI were compared. We then evaluated the difference between MEP amplitude and functional assessment according to FN and FA of DTI. ROC curves were used to determine the optimal cutoff score for the affected side’s FN and FA of DTI using MEP amplitude and functional assessment. The correlation between FN and FA of DTI, and the MEP amplitude and functional assessment was analyzed.
Results
MEP amplitude and functional assessment were different between the affected and non-affected sides. FN and FA of DTI were also different between the affected and non-affected sides. When we set FN as 70 and FA as 0.4, FDI and TA amplitude in MEP studies and HFTs [grip strength, Box and Block (B&B) test, nine-hole peg test (NHPT)] showed significant differences between MEP amplitude and functional assessment.
We determined meaningful optimal cutoff values by setting sensitivity, specificity, and area under the curve to ≥70%. FN of CST was 64 among patients with evoked amplitude in MEP, 142.5 among patients with ankle dorsiflexion MRC scale ≥ grade 2, 178.5 among patients with finger flexion MRC scale ≥ grade 2, 178.5 among patients with measurable grip strength, and 197 among patients who could perform NHPT.
FA of CST was 0.555 among patients with evoked amplitude in MEP, 0.575 among those with finger flexion MRC scale ≥ grade 2, and 0.595 among those with measurable grip strength and who could perform the B&B test.
FN and FA of DTI were correlated with MEP amplitude and functional assessment.
Conclusion
We conclude that FN and FA of DTI may be useful in predicting functional motor recovery in patients with subacute brain injury.
|
|
|
