| BackgroundIntramedullary high signal intensity on T2-weighted magnetic resonance images (MRI) in cervical spondylotic myelopathy may reflects some kind of pathological state of the spinal cord with chronic compression, but the specific clinical significance remains controversial. In this study, the first part of the results of clinical studied show that,There was a statistically significant association between DTI values of MRI T2WI high intensity signal of cervical spinal cord and spinal cord functional recovery indicators, such as JOA Recovery Rate and postoperative JOA score, which are valuable predictors for the clinical outcome of surgical decompression for CSM. And the main theoretical basis is that DTI may reflect subtle pathophysiological changes of the cervical spinal cordin CSM. but all these are only clinical inference, and lack of pathological evidence. So it is important to create a cervical spondylotic myelopathy animal model to study the characteristics of DTI values of MRI T2WI high intensity signal of cervical spinal cord and corresponding pathological changes, for providing pathological theory for clinical application of DTI values.Objectives To study the Pathological Mechanism and corresponding MRI DTI of Cervical spinal cord in Goat Model with Cervical Spondylotic Myelopathy, through establishing a goat model with chronic cervical spinal cord compression, for providing pathological theory for clinical application of DTI values.Methods and materials1. Four Healthy goats, weight from20kg-25kg.Right C4/5laminar space was explored, and part of the right C4laminectomy was performed by rongeur, and one balloon was inserted into the epidural space slowly and carefully, up to the right side of the front at C4vertebral body level. From2weeks postoperation,0.05ml contrast agent were injected into the balloon through the catheter slowly every twenty days, and total volume was less than0.3ml, for establishing a goat model with chronic cervical spinal cord compression.2. Spinal cord motor function was scored by the modified Tarlov criteria for the two goat models with chronic cervical spinal cord compression.3. Pathology evaluation:two goat models were given a lethal injection of potassium chloride, and3cm longer spinal cord tissue which include the compressed lesion for pathology evaluation after MRI DTI12W and24W postoperation respectively.4. MRI DTI scans were performed on four goats preoperation, and on two goat models postoperation. SIEMENS MAGNETOM Verio3.0T Magnetic resonance imaging (MRI) studies with T1-and T2-weighted in sagittal/axial scans were undertaken for all patients and volunteers.A eight-channel NV-full coil was accepted. All the data of DTI were analyzed and imagings were reconstructed with NEURO3D software at SIEMENS working station. Conventional MRI scan sequence for cervical spinal cord:Goats wer ere scanned with a fast gradient echo sequence, and T1WI, T2WI and cross-sectional T2WI scan of cervical spinal cord was obtained using a single-shot fast spin-echo-based sequence at3.0T. Scanning parameters:Sagittal T1WI (TR/TE,600/15ms)、 T2WI (TR/TE,5200/140ms); and T2WI (TR/TE/IR,9000/120/2100ms), thickness3mm, spacing of1.0mm。 T1coronal TLC scanning, thickness1.0mm, spacing0,160layer for integration with DTI. and FOV210×210mm, Matrix512×512。Results1. Four goats accepted operation for establishing model, but two goats failed due to acute spinal cord injury.12weeks after operation, goats were injected with0.2ml contrast agent into the balloons, and the function score for standing and walking motor is5. MRI shows the balloons were oval, the right side of the cervical spinal cord were compressed. There is a slight deformation on DTI images with no obvious artifacts. There was a punctuate T2high signal and T1normal signal on the compression lesion.24weeks after operation, the last goat was injected with0.3ml contrast agent into the balloons, and the function score for standing and walking motor is only3.5. MRI shows that patchy T2high intensity signal, wider than3months ago, and T1normal signal.DTT imaging:24weeks after operation, there is a slight deformation on DTI images with no obvious artifacts.DTT show mild deformation on the trace fiber bundles with no obvious fracture and destruction.24weeks after operation, DTI imaging shows that slight deformation which is more severe than that3months ago. FA imaging show the color of spinal cord decline than preoperative one, and There was a obvious pressure trace made by balloon, and the fiber bundle become thinning and twist, with no obvious fracture signs.DTI values:the mean value of preoperative FA of cervical spinal cord is0.763in goats, and ADC1.040×10-3mm2/s, MD is similar with ADC, λ‖is2.155×10-3mm2/s and λ⊥is0.482×10-3mm2/s。12weeks after operation, the change trend of DTI values of the cervical spinal cord is the same:preoperation FA is lower than preoperative FA, and ADC、λ‖、λ⊥and MD were all increased.24weeks after operation, FA is decreased significantly, and ADC、λ‖、λ⊥and MD were all increased significantly.Pathological results:12weeks after operation, most neuronal morphology is generally mormal, a little spinal cord gray matter neurons pyknosis, eossinophilic change. Spinal cord white matter showed myelin edema, edema of the nerve fibers, Wallerian degeneration, we can see a small amount of cystic area.24weeks after operation, there is unclear boundaries between gray and white matter, and the proliferation of fiber bundles interspersed at meantime, neurons arranged clutter, a small mumber of neurons in the cell is still visible pyknotic and eosinophilic change. White matter edema is not obvious, part of destructed nerve fibers recovery to normal, but there are still some obvious cystic degeneration, and red dye substances.Conclusions1. Goat models with chronic cervical spinal cord compression were established successfully.2. The change trend of DTI values of chronic compressed cervical spinal cord is, with the aggravation of spinal cord compression, ADC、λ‖、λ⊥and MD are increasing, and FA is decreasing.3. Unlike acute spinal cord injury, this study indicates that pathologic change in the chronic compression cervical spinal cord contains both reversible and irreversible pathological changes. It is impossible to be reversible or irreversible pathological changes only. BackgroundCervical spondylosis myelopathy is the most serious condition that can result from cervical spondylosis, and is the most common cause of progressive spinal cord dysfunction among those aged over55years all over the world. It is important to have early, accurate diagnosis and assessment of cervical spinal cord injury for the treatment and prognosis of cervical spondylotic myelopathy. Intramedullary high signal intensity on T2-weighted magnetic resonance images (MRI) in cervical spondylotic myelopathy may reflects some kind of pathological state of the spinal cord with chronic compression, but the specific clinical significance remains controversial.MRI diffusion imaging is the only way to measure the diffusion movement and imaging of water molecules in vivo, among that DWI is the best way to observe the water diffusion in vivo, and DTI technology developed on the basis of which can quantify the directionality of the diffusion of water molecules by the use of a number of different directions diffusion, and measure the value of values of apparent diffusion coefficient (ADC), fractional anisotropy (FA), and perform diffusion tensor tracking (DTT). To provide objective measures in terms of microstructural change of spinal cord in vivo for clinical diagnosis. Due to the special anatomic characteristics, such as small size spinal cord and bony spinal, the use of DTI in the cervical spinal cord was restricted. In recent years, with the rapid development of magnetic resonance technology, DTI has been applied to the study of spinal cord lesion, but most of which are based on1.5TMRI. Few authors take DTI quantitative research for CSM by the use of3.0T MRI, and found that ADC and FA values is a sensitive indicator for the evaluation of water diffusion anisotropy and strength, to make up for the shortcoming of conventional MRI for not able to quantitative analysis of spinal cord lesions. The value of λ1、λ2andλ3can evaluate the diffusion strength of water molecules in the cervical spinal cord. DTT imaging provide visualization of the anatomical structure of the white matter fiber tracts in the cervical spinal cord and nerve roots, contribute to non-invasive detection of the destruction of the cervical spinal cord nerve fibers. But there is still lack of study for the correlation between DTI quantitative indicators with the evaluation predictor for the function recovery of spinal cord before and after surgical decompression for CSM, and no further study of the predictive significance of DTI value for MRI T2WI high signal in the spinal cord of CSM, And lack of corresponding pathology for MRI T2WI high signal intensity of spinal cord.so, it is helpful for predicting the clinical outcome to know the pathological change of the cervical spinal cord lesion through DTI quantitative check on3.0T MRI. It is important to investigate the correlation between the DTI values and function recovery of spinal cord before and after surgical decompression, and make its corresponding pathological changes clear. Therefore, this study obtain the values of ADC, FA, λ1、λ2andλ3, combined with DTT imaging for spinal cord for high intensity signal of CSM before and after surgical decompression, with3.0T MRI, and study the correlation between the DTI values and function recovery of spinal cord before and after surgical decompression, in order to provide theory for clinical practice.ObjectivesComparisons of DTI values of spinal cord between the three groups, CSM patients with and without MRI T2WI high intensity signal, and normal volunteers, to explore the characteristics of DTI values of MRI T2WI high intensity signal of CSM.Further clarify the characteristics of DTI values of spinal cord with chronic compression of CSM through comparisons of DTI values before and after surgical decompression, to investigate the corresponding pathological changes of DTI values of MRI T2WI high intensity signal of CSM.Study the correlation between DTI values of spinal cord with chronic compression of CSM before and after surgical decompression with JOA recovery rate, to investigate the predictive value of DTI of of MRI T2WI high intensity signal of CSM.MethodsThis study was approved by an institutional review board. Informed consent was obtained from all patients and volunteers. Patients who experienced CSM and underwent subsequent cervical spinal cord decompression and reconstruction were selected in the Guangzhou First People’s Hospital, from January,2010through December,2012.(According to literature, α=0.05,1-β=0.8, μA=0.85, μB=1.15, σ=0.58,△=μA-μB/σ, Number of patients=16/△2=59)Inclusion and exclusion criteria:(1) with Clinical manifestations and signs of spinal cord injury,(2)Confirmed compression of cervical spinal cord in MRI,(3) Cervical spine trauma, tumor, motor neurone disease, peripheral neuritis and so on were excluded.Grouping, CSM patients were grouped according to the signal change of cervical spinal cord between preoperative and postoperative MRI, based on the current consensus about the high intensity signal of spinal cord:bad clinical outcome for CSM with T2WI high intensity signal preoperative combined with preoperative or postoperative T1WI lower intensity signal. There four study groups:G0group, control group with health volunteers; G1group, CSM patients without MRI T2WI high intensity signal of cervical spinal cord; G2group, CSM patients with preoperative MRI T2WI high intensity signal, combined with preoperative or postoperative T1WI lower intensity signal; G3group, CSM patients with preoperative MRI T2WI high intensity signal, and which decreased or disappear after surgical decompression.Clinical data, including name, age, sex, comorbidities, duration, surgical decompression, blood loss, operative time, perioperative complications and so on were recorded detail.Functional AssessmentThe neurological condition of patients was assessed using the Japanese Orthopaedic Association myelopathy score (JOA score)(130). Data were recorded in the medical record before surgery and in the immediate postoperative period, and the recovery rate proposed by Hirabayashi et al. was also used for neurological assessment: Recovery rate (%)=(postoperative JOA score-preoperative JOA score)/(17-preoperative JOA score)×100. Results were indicated by the recovery rate as follows:75%or more (excellent),50%to74%(good),25%to49%(fair), and less than25%(poor).Radiological evaluationPreoperatively, anteroposterior, oblique, neutral lateral and lateral flexion-extension X-ray studies of the cervical spine were undertaken for all patients. Further anteroposterior and neutral lateral X-rays were performed within the first postoperative week,6month,12month,24month postoperatively. From the lateral radiograph, cervical lordosis or kyphosis was measured at each level from C2to C7according to the Ishihara Index (IshI)(135):(al+a2+a3+a4) mm/A mm×100. The change in Ishihara index was defined by the difference in value between postoperative cervical curvature indices and preoperative ones, and loss of balance (toward kyphotic deformity) was defined by a reduced IshI of at least10on the last acquired lateral X-ray study. SIEMENS MAGNETOM Verio3.0T Magnetic resonance imaging (MRI) studies with T1-and T2-weighted in sagittal/axial scans were undertaken for all patients and volunteers.A eight-channel NV-full coil was accepted. All the data of DTI were analyzed and imagings were reconstructed with NEURO3D software at SIEMENS working station. Further same MRI examinations were performed at the last follow up postoperatively.Conventional MRI scan sequence for cervical spinal cord:All patients and volunteers were scanned with a fast gradient echo sequence, and T1WI, T2WI and cross-sectional T2WI scan of cervical spinal cord was obtained using a single-shot fast spin-echo-based sequence at3.0T.Image processing and data measurement and assessment:DTI values (include values of ADC, FA,λ1、λ2and λ3) of high intensity signal were obtained and DTT imaging were reconstructed.Statistical MethodsFor each pre-, peri-, and postoperative variable, not only was an overall probability value generated via groups’ comparison, but individual group comparisons were performed as well. For categorical variables (sex, presence of complication, presence of comorbidity), these comparisons were achieved using chi-square or Fisher exact tests. For continuous data (age, duration from onset of symptoms to surgery, pre-operative JOA score, post-operative JOA score, pre-operative Ishihara index, post-operative Ishihara index, operative blood loss, operating room time, number of complications, JOA Recovery Rate, change in Ishihara index, DTI values include values of ADC, FA, λ1、λ2and λ3), the one-way ANOVA or Independent-Sample T-test was used.Multiple linear regression modeling with adjustment for all observed covariates (age, sex, operative approach, presence of comorbidities, duration from onset of symptoms to surgery, pre-operative JOA score, pre-operative Ishihara index) was performed to examine how they affected the targeted outcomes of JOA Recovery Rate and postoperative JOA score. Such analysis can stringently control for confounding variables among the covariates and is hence expected to yield less profound results than the1-way ANOVA or Independent-Sample T-test. This technique permits estimation of regression coefficients and associated confidence intervals for the desired outcome variables.The level of significance was defined as0.05. Statistical analyses were performed using SPSS17.ResultsWe identified a total of61consecutive patients age form41years to92years (mean age62.41±12.25years) who experienced CSM and underwent subsequent cervical spinal cord decompression and reconstruction from January,2010through December,2012.39patients were male, and22were female. All the patients were followed up14months to26months (mean time17.8months).20consecutive healthy qualified volunteers in the normal control group over the same period, age form39years to78years, with an average age of59.40±9.84years.Sex comparisons of patient characteristics between G1versus G0, G2versus GOand G3versus GO did not vary significantly (χ2=2.042, P=0.564). and age comparisons of patient characteristics between G1versus G0, G2versus GOand G3versus GO did not vary significantly (F=3.379, P=0.07)Each of the patient cohorts did not significantly differ from each other in terms of comorbidities, complications or operative approach (P>0.05). There also was no significant difference between the four groups regarding duration from onset of symptoms to surgery (F=6,807, P=0.004), as mean duration from onset of symptoms to surgery were3.70±4.52months in the G1group,22.95±21.98months in the G2group and9.95±14.71months in the G3group. Patients experienced the significant shortest duration from onset of symptoms to surgery, compared with G2or G3group respectively (F=-3.231,P=0.006; F=-2.05,P=0.049)There also was no significant difference between the four groups regarding preoperative JOA score (F=0.725, P=0.489). The mean postoperative JOA score were15.10±1.34in the G1group,11.00±1.80in the G2group and14.23±1.88months in the G3group. Patients experienced the lowest preoperative JOA score in G1group, compared with G2or G3group respectively (t=7.730, P=0.000; t=-5.257, P=0.000) and there was no significant difference between the G1groups and G3groups regarding postoperative JOA score (t=1.772, P=0.083).All the patients in the G1、 G2and G3group experienced improvement of JOA recovery rate G1、G2、G3, There was no significant difference between the three groups regarding postoperative JOA score(F=16.728, P=0.000). The mean JOA recovery rate were70.24%in the G1group,59.71%in the G2group, and There was no significant difference compared with that in the G3group(9.48%) respectively (t=5.840,P=0.000; t=-5.152,P=0.000). There was no significant difference between G1and G3group regarding JOA recovery rate (t=1.162,P=0.114)Fractional anisotropy (FA)The mean preoperative FA were0.665±0.455in the G1group,0.422±0.748in the G2group and0.586±0.591months in the G3group. And the mean preoperative FA were0.597±0.478in the G1group,0.407±0.735in the G2group and0.548±0.106months in the G3group. preoperative FA in G1group is higher than that in GOgroup (t=-4.579, P=0.000). And Patients experienced the lower preoperative FA in G2or G3group, compared with GO group respectively (t=9.157, P=0.000; t=2.082, P=0.045). There were no significant difference between the G1、 G2and G3groups regarding preoperative FA and Postoperative FA respectively (F=71.930, P=0.000; F=22.620, P=0.000). Further comparison, patients experienced higher preoperative FA in G1, compared with that in G2and G3group respectively (t=11.957, P=0.000; t=5.050, P=0.000). patients experienced higher preoperative FA in G3, compared with that in G2group (t=-7.601, P=0.000)Patients experienced higher postoperative FA in G2, compared with that in G1and G3group respectively (t=9.256, P=0.000; t=-4.370, P=0.000). patients experienced higher preoperative FA in G3, compared with that in G2group (t=1.980, P=0.054) Apparent diffusion coefficient (ADC) and MD value. The mean preoperative ADC were1.165±0.146、1.641±0.147、1.362±0.166in the G1, G2and G3group respectively, and the mean preoperative ADC were1.355±0.154、1.857±0.247、1.415±0.386in the G1, G2and G3group respectively. The mean preoperative MD were1.166±0.146、1.641±0.147、1.362±0.166in the G1, G2and G3group respectively, and the mean preoperative MD were1.355±0.154、1.874±0.234、1.415±0.386in the G1, G2and G3group respectively. ADC value is similar with MD value. Compared with ADC and MD value in GO group, that in G1is a little bit lower (P<0.05), and those in G2and G3group were much higher (P<0.05). There was no significant difference between the three groups regarding preoperative ADC and MD. Preoperative ADC and MD in G2group were much higher than those in G1or G3group (P<0.05); and that in G3group is higher than that in G1group (P<0.05) Postoperative ADC and MD in G2group were also much higher than those in G1or G3group(P<0.05), but there was no significant difference between G1and G3group regarding postoperative ADC and MD. So, change trend of ADC and MD value is consistent.6. λ‖and λ⊥valueThe mean preoperative λ‖were2.215±0.320、2.392±0.194、2.387±0.271in the G1, G2and G3group respectively, and the mean postoperative λ‖were2.391±0.226、2.776±0.290、2.353±0.421in the G1, G2and G3group respectively. There was significant difference between G1and GOgroup regarding preoperative λ‖(t=4.146, P=0.000). Patients experienced the higher preoperative λ‖in G2or G3group, compared with GO group respectively (t=-9.924, P=0.000; t=-2.638, P=0.011). There was no significant difference between the three groups regarding preoperative λ‖(F=2.735, P=0.073), but There was significant difference between the three groups regarding postoperative λ‖(F=7.958, P=0.001). Patients experienced the lower postoperative λ‖in G1or G3group, compared with G2group respectively (t=-4.405, P=0.000; t=3.348, P=0.000). there was significant difference between G1and G3group regarding postoperative λ‖(t=0.386, P=0.714)There was significant difference between G1and GOgroup regarding preoperative λ⊥(t=0.370, P=0.713). Patients experienced the higher postoperative λ⊥in G2or G3group, compared with GO group respectively (t=-2.532, P=0.016; t=-2.860, P=0.007).There also were significant differences between preoperative λ⊥and postoperative λ⊥among G1, G2and G3group (F=7.958, P=0.001; F=87.575, P=0.000; F=30.40, P=0.000). Patients experienced the lower pretoperative λ⊥in G1or G3group, compared with G2group respectively (-11.988, P=0.000; t=8.040, P=0.000). Patients experienced the higher pretoperative λ⊥in G3group, compared with G1group (t=-5.826, P=0.000). Patients experienced the lower pretoperative λ⊥in G1or G3group, compared with G2group respectively (t=-8.853, P=0.000; t=4.137, P=0.000). there was significant difference between G1and G3group regarding postoperative λ⊥(t=-1.232, P=0.224)There was a statistically significant association between DTI values of MRI T2WIhigh intensity signal of cervical spinal cord and spinal cord functional recovery indicators, such as JOA Recovery Rate and postoperative JOA score at a5%level of significance after adjusting for other demographic, neurologic assessment and spinal canal radiologic evaluation variables, and there was no statistically significant association between DTI values of MRI T2WIhigh intensity signal of cervical spinal cord and preoperative JOA score at a5%level of significance.Preoperative FA was positively correlated with JOA recovery rate and postoperative JOA score[0.54(99.05,246.61);0.54(6.944,15.900)], Preoperative ADC was passively correlated with JOA recovery rate and postoperative JOA score[-0.55(-0.11,-0.05);-0.52(-0.007,-0.003)], Preoperative MD was passively correlated with JOA recovery rate and postoperative JOA score[-0.55(-0.11,-0.05);-0.52(-0.007,-0.003)], Preoperative λ‖was passively correlated with JOA recovery rate and postoperative JOA score[-0.30(-0.07,-0.01);-0.28(-0.004, 0.000)],Preoperative λ⊥was passively correlated with JOA recovery rate and postoperative JOA score[-0.60(-0.11,-0.05);-0.598(-0.007,-0.003)].Conclusion1. Compared with normal control group, FA in the G1group is significantly higher, and ADC, λ1、λ2andX3were significantly lower. FA in the G2group is significantly lower, and ADC, λ1、λ2andλ3were significantly higher. The same MRI T2WI high signal spinal cord with G2group, FA in the G3group is relatively lower, and ADC,λ1、λ2andλ3were relatively higher.2. MRI T2WI high intensity signal of cervical spinal cord may represent two completely opposite pathological state, and which is difficult to distinguish by observing the high signal itself, but DTI values can evaluate the pathology of the MRI T2WI high intensity signal at the pathological molecular level. So the DTI values are valuable predictor for the clinical outcome of surgical decompression for CSM, and are helpful for the development of surgical plan.3. Preoperative and postoperative DTI values are correlated with postoperative JOA score significantly, which indicate that DTI values can reflect the pathological state of compressed cervical spinal cord. Most important, There was a statistically significant association between DTI values of MRI T2WI high intensity signal of cervical spinal cord and spinal cord functional recovery indicators, such as JOA Recovery Rate and postoperative JOA score。4. DTT imaging provides visualized detection of the destruction of the cervical spinal cord nerve fibers. Which are valuable supplement of DTI values for predicting the clinical outcome of CSM. |
PDF Full Text Request