The Application Of Diffusion Tensor Imaging In Early Assessment Of Prognosis Among Patients With Limb Motor Dysfunction After Cerebral Infarction

Objective: Dynamical observation of the DTI parameter changes incerebral infarction at different clinical stages, and observed the morphologicalchanges of the corticospinal tract (CST) by using the diffusion tensortractography (DTT) technology to evaluate the extent of motor dysfunctionand clinical prognosis on primary stage.Methods: From October2010to December2011,30cases of patients(male18, female12, mean age62.44±14.96years old, range29to80yearsold) presenting within72hours of the onset were studied. All patients sufferedvarying degrees of unilateral limb movement dysfunction, in addition theresponsible focus was located in the internal capsule, the lateral ventricle sideof body, or the corona radiata area. All patients were in line with acuteischemic stroke treatment guidelines. At the time of acute phase (within72hours), subacute phase (4weeks) and chronic phase (12weeks), NationalInstitutes of Health Stroke Scale (NIHSS) was used to evaluate the impairmentextent of neurologic function respectively. In acute phase, patients weredivided into two groups as one of the prognosis indice: recovering bettergroup (NIHSS<7point) and recovering worse group (NIHSS≥7point).Modified Rankin Scale (mRS) was used to evaluate the damaged extent ofmotor function in patients. MRS was divided into6grades, grade0¹: normalor approaching to normal, grade2: mild disability, grade3: moderate disability,grade4: moderatesevere disability and grade5: severe disability. MRS<3wasconsidered as recovery group and mRS≥3was considered as paralysis group.Conventional MRI, diffusion weighted imaging (DWI) and DTI wereperformed on a Siemens Tim-avanto1.5T in acute phase, subacute phase andchronic phase. According to selected region of interest (ROI) in factional anisotropy (FA) map, apparent diffusion coefficient (ADC) map and FA colorcoded map, the values of FA and ADC in the infarction region, thecorresponding region and the cerebral peduncle region were measured. Andthen the relative FA (rFA) and relative ADC (rADC) were calculated. BilateralCST that were performed by using DTT technology could be divided into3grades according to its degree of damage. The grade0of CST was complete,grade1was complete but compression shift and grade2was break. Patientswere divided into two groups: recovering better group (grade0and grade1)and recovering worse group (grade2) in acute phase as the prognosis indice.In chronic phase, mRS score was regarded as the standard to compare theprecision of NIHSS score and the grade of CST about evaluating prognosis.The statistics analysis was made by SPSS13.0and P<0.05was considere-d to having significant changes.Results:1General materials: The during time that30patients were suffered from thecerebral infarction was range6to48hours, the average time was26±17hours.All the lesions could be seen on conventional MRI, such as10cases located inthe internal capsule,7cases in corona radiate,9cases in lateral body and4cases in semioval center. Some patients merged more than one risk factors,such as21cases with hypertension,11cases with diabetes,6cases withdyslipidemia,8cases with coronary heart disease,13cases smoking and5cases with the history of alcohol abuse.2Neurological function scores2.1The NIHSS score: In actue phase, there were19cases in NIHSS<7group and11cases in NIHSS≥7group. All patients' NIHSS score ranged form0to13points in acute phase, a median of5points. In subacute phase, theNIHSS score ranged from0to10points, a median of2.5points. In chronicphase, the NIHSS score ranged from0to8points, a median of1points. TheNIHSS scores in different periods were compared with significant change bynon-parametric Wilcoxon rank sum test(χ²=12.035, P<0.05). 2.2The mRS score: In chronic phase, patients were divided into6group:14cases of grade0,8cases of grade1,3cases of grade2,3cases of grade3,2cases of grade4and0cases of grade5.3The parameter variation of the DTI in different phases (the unit of theADC value:10^-2mm²/s)3.1Acute phase:(1) Comparing the value of FA between the infarctionregion and contralateral corresponding region, the change was significant(t=-8.805, P<0.05). In the cerebral peduncle region, there was significantdifference in FA between two sides (t=-3.011, P<0.05).(2) The value of ADCbetween the infarction region and contralateral corresponding region hadstatistical significance (t=-9.811, P<0.05). In the cerebral peduncle region, thevalue of ADC between two sides had no statistical significance (t=-1.134, P＞0.05).3.2Subacute phase:(1) Comparing the value of FA between the infarctionregion and contralateral corresponding region, the change was significant(t=-5.411, P<0.05). In the cerebral peduncle region, difference of the value ofFA between two sides was significant (t=-5.411, P<0.05).(2) The value ofADC between the infarction region and contralateral corresponding region hadstatistical significance (t=-2.532, P<0.05). In the cerebral peduncle region, thevalue of ADC between two sides had no statistical significance (t=-1.745, P＞0.05).3.3Chronic phase:(1) Comparing the value of FA between the infarctionregion and contralateral corresponding region, the change was significant(t=-5.289, P<0.05). In the cerebral peduncle region, difference of the value ofFA between two sides was significant (t=-5.289, P<0.05).(2) The value ofADC between the infarction region and contralateral corresponding region hadstatistical significance (t=2.713, P<0.05). In the cerebral peduncle region, thevalue of ADC between two sides had no statistical significance (t=-0.932, P＞0.05).3.4In the infarction region, the value of FA had statistical significance(F=8.693, P<0.05), and the value of ADC had statistical significance (F=48.96,P<0.05). In the cerebral peduncle region, the change of the value of FA was significant (F=9.86, P<0.05), but the value of ADC had no statistical signifi-cance.4The relationship between the infarction and CSTIn acute phase, there were6cases classified into grade0,15cases intograde1and9cases into grade2. In subacute phase, there were13casesbelong to grade0,8cases belong to grade1and9cases belong to grade2. Inchronic phase, the number of grade0was13, the number of grade1was5andthe number of grade2was9.5The relationship among the DTI parameter, the grade of CST and neurolo-gyical function scores5.1In acute phase, the value of rFA in the infarction region showed thenegative correlation among the different periods (acute phase, subacute phaseand chronic phase) of NIHSS score (r₁=-0.515, r₂=-0.505, r₃=-0.450P<0.05).There was the negative correlation between the value of rFA in the infarctionregion and mRS score (r=-0.449, P<0.05).5.2In subacute phase, the value of rFA in the cerebral peduncle regionassumed the negative correlation among the different periods of NIHSS score(r₁=-0.654, r₂=-0.654, r₃=-0.674P<0.05). There was the negative correlationbetween the value of rFA in the cerebral peduncle region and mRS score(r=-0.614, P<0.05).5.3The grade of CST showed the positive correlation among the differentperiods of NIHSS score (r₁=0.654, r₂=0.654, r₃=0.674P<0.05). There was thepositive correlation betweer the grade of CST and mRS score(r=0.832,P<0.05).6The sensitivity, specificity and overall accuracy rate of prediction of thegrade of CST were80%,80%and80%. And that of the NIHSS score were72%,80%and73%. There was significant difference in the accuracy rate ofpredictive value between the two sides (P<0.05).Conclusion:1The values of FA and ADC in different phases of cerebral infarction haveobviously variation trend. According to the dynamic changes of DTI parameter, FA and ADC values can be considered as one of the objectiveevidence to judge the clinical effects.2The value of rFA in the infarction region and the grade of CST, as anobjective index, not only can analyse the damage of motor function, but alsoassess the recovery extent of motor function at early stage.3The accuracy rate of predictive value in the grade of CST is better thanthat of NIHSS score.