| BackgroundsGuillain-Barre syndrome(GBS)is one of the most common type of autoimmune neuropathy,which expressed as the pathological changes of peripheral nerve inflammatory demyelination.Some patients even found to have limb paralysis,respiratory paralysis or some other life-threatening symptoms.The recovery of the peripheral nerve is slow and it is difficult to reliably assess the damage of peripheral nerve by clinical examination,electrophysiological examination and routine imaging examination.As a kind of physiological examination,the electrophysiological examination is highly sensitive to abnormally nerve conduction,but it is invasive,subjective and the results may be affected by potential biochemical abnormalities(such as blood calcium,potassium,magnesium ion concentration)and drugs.Sural nerve biopsy is an auxiliary means of inspection,demyelination and inflammatory cell infiltration may suggest GBS,but the sural nerve is sensory nerve and GBS mainly affected the motor nerves,so the sensitivity and specificity of the results can only be used as reference.Therefore,the diagnosis of GBS requires an objective,non-invasive and accurate method for assessing the injuries of peripheral nerves.It is important to display the peripheral nerve morphology,movement and pathological changes clearly,directly and non-invasively,which has also been the focus and difficulty of imaging diagnosis.High resolution ultrasound can show the anatomical structure of superficial nerves,with real-time observation,high spatial resolution and low cost,and dynamic observation of nerve movement is the unique advantage;however,it is subjective and difficult to display the nerves located in deep position.Therefore,the examination is limited in the diagnosis of peripheral nerve injuries in GBS.Magnetic resonance imaging(MRI)is increasingly applied to the study on imaging and pathological changes of peripheral nerve because of its excellent soft tissue resolution and multi-directional,multi parameters and multi sequence imagings.Magnetic resonance neurography(MRN)can quantitatively evaluate the integrity,pathological changes and relationships of structure and function for peripheral nerves in cellular and molecular level,which can provide more accurate diagnosis and treatment for clinical basis.Now DWI and T2 fat suppression technology are the most commonly used MRN techniques.Diffusion tensor imaging(DTI)and background suppression of diffusion weighted imaging(DWIBS)are the new MRN technologies which developed on the basis of DWI.The purpose of this study was to investigate the value of MRN in evaluating GBS peripheral nerve injuries by analyzing the DTI and DWIBS features of the tibial nerve(TN)and common peroneal nerve(CPN)in patients with GBS.Part Ⅰ Quantitative study of diffusion tensor imagingof the tibial and common peroneal nerve in patients with Guillain-Barre syndromeObjectiveThe main objective of this part is to explore the statistical difference in apparent diffusion coefficient(ADC),fractional anisotropy(FA),axial diffusivity(AD),radial diffusivity(RD)value of TN and CPN between normal volunteers and Guillain-Barre Syndrome(GBS);to detect correlations between the parameters of DTI and electrophysiology parameters.Materials and Methods1.Study populationThirty six lower extremities of 20 GBS patients(13 men,7 women,mean age:40 years,age range 14 to 71 years)and thirty two lower extremities of 16 healthy volunteers were examined in this study.The GBS patients and healthy volunteers were sex and age matched.Electrophysiological examination showed decrease of MCV in different degrees and(or)decreased motor nerve conduction amplitude in lower limb.In the 20 GBS patients,there were 10 patients with cerebrospinal fluid protein cell separation,10 cases of cerebrospinal fluid protein content had no obvious increase.2.Electrophysiology examinationWithin one week before the MRI examination,all GBS patients underwent electrophysiology of the lower limbs by experienced examiners.The results were independently assessed by a board-certified clinical neurophysiologist.MCV and motor nerve conduction amplitude of each subject were recorded.The normal MCV value is≥41m/s in TN and CPN.The normal amplitude values are ≥4mV in TN and>2mV in CPN.3.MR examinationAll subjects were examined using a commercially available clinical 3T MRI system(Achieva,Philips Healthcare,Best,the Netherlands).An eight-channel knee coil was employed at the knee level.The subjects were placed supine with feet-first position and unilateral imaging.Conventional MRI sequences included axial and coronal T1 weighted imaging(T1WI):repetition time/echo time(TR/TE)550/20ms,field of view(FOV)120mm×120mm,slice thickness 3.0mm,overlap 0.3mm,number of excitations(NEX)2,number of slices 45.Axial and coronal T2 weighted imaging spectrally selective attenuated inversion recovery(T2WI SPAIR):TR/TE 3000/55ms,FOV 120 mm×120mm,slice thickness 3.0mm,overlap 0.3mm,NEX 2.DTI:TR/TE 4892/103ms,FOV 160 mm×160mm,image matrix 128×125,slice thickness 3.0mm,overlap 0,sensitivity-encoding factor 2.3,number of directions of motion-probing gradients 15.b-value 800s/mm2,NEX 2.4.DTI post-processingThe raw diffusion tensor data were processed on an independent workstation(Extended MR Workspace,Philips Healthcare).T1WI images were used for image fusion with color FA diagram.ROIs were placed manually on TN and CPN respectively,through the center of knee joint,on the upper 10mm and lower 10mm slices respectively.ADC value and FA value were automaticly generated.The average measurements of the FA and ADC values in the three ROIs on tensor calculated images were recorded by two radiologists independently,and the average of the results were used for further analysis.Mricron,DtiStudio and Image J were used to measure the AD value and RD value at the same level.MR images were analyzed by two independent radiologists,blinded to each other.5.Statistical analysisThe ADC,FA,AD and RD value of tibial nerve and common peroneal nerve in the GBS group and the normal group were examined by the normal distribution.The inter-reader agreement of DTI parameters were assessed using intraclass correlationcoefficient(ICC)analysis.The ADC,FA,AD and RD value of the two groups were compared with the independent sample t test,and p<0.05 was statistically significant.Receiver operating characteristic(ROC)analysis was performed to evaluate the diagnostic performance of DTI parameters for TN and CPN.Pearson correlation coefficient was performed for the analysis of the correlation between DTI parameters and electrophysiology parameters.Statistical analysis was performed using SPSS software(version 19).Results1.The inter-reader agreement of DTI parameters of TN and CPN ICC for inte-reader agreement of ADC,FA,AD and RD measurements were 0.883(95%CI,0.746-0.941),0.918(95%CI,0.862-0.959),0.892(95%CI,0.819-0.938),0.860(95%CI,0.724-0.916)of TN respectively,and 0.895(95%CI,0.752-0.963),0.925(95%CI,0.869-0.974),0.857(95%CI,0.809-0.914),0.873(95%CI,0.761-0.942),respectively.The inter-reader agreement of DTI parameters of TN and CPN was good.2.Comparison of ADC values between GBS group and healthy volunteers The ADC values of TN in GBS group and healthy volunteers were(1.31±0.16)×10-3 mm2/s,(1.10±10.12)×10-3 mm2/s,respectively.The difference was statistically significant(p<0.05).Using 1.20×10-3 mm2/s as the threshold,which was obtained by ROC analysis,the diagnostic sensitivity,specificity,and aera under the curve(AUC)were 77.8%,82.4%,0.865,respectively.The ADC values of CPN in GBS group and healthy volunteers were(1.23±0.17)×10-3 mm2/s,(1.03±0.12)×10-3 mm2/s,respectively.The difference was statistically significant(p<0.05).Using 1.06×10-3 mm2/s as the threshold,which was obtained by ROC analysis,the diagnostic sensitivity,specificity and AUC were 77.8%,82.4%,0.865,respectively.3.Comparison of FA values between GBS group and healthy volunteersThe FA values of TN in GBS group and healthy volunteers were 0.51±0.05,0.60±0.04,respectively.The difference was statistically significant(p<0.05).Using 0.57 as the threshold,which was obtained by ROC analysis,the diagnostic sensitivity,specifiicity and AUC were 88.9%,100%,0.970,respectively.The FA values of CPN in GBS group and healthy volunteers were 0.53±0.06,0.63±0.03,respectively.The difference was statistically significant(p<0.05).Using 0.58 as the threshold,which was obtained by ROC analysis,the diagnostic sensitivity,specificity and AUC were 88.9%,94.1%,0.927,respectively.4.Comparison of AD values between GBS group and healthy volunteersThe AD values of TN in GBS group and healthy volunteers were 0.0011±0.0005,0.0020±0.0003,respectively.The difference was statistically significant(p<0.05).Using 0.0013 as the threshold,which was obtained by ROC analysis,the diagnostic sensitivity,specificity and AUC were 77.8%,100%,0.929,respectively.The AD values of CPN in GBS group and healthy volunteers were 0.0010±0.0004,0.0018±0.0003,respectively.The difference was statistically significant(p<0.05).Using 0.0011 as the threshold,which was obtained by ROC analysis,the diagnostic sensitivity,specificity and AUC were 75.0%,100%,0.900,respectively.5.Comparison of RD values between GBS group and healthy volunteersThe RD values of TN in GBS group and healthy volunteers were 0.0007±0.0002,0.0010±0.0002,respectively.The difference was statistically significant(p<0.05).Using 0.0008 as the threshold,which was obtained by ROC analysis,the diagnostic sensitivity,specificity and AUC were 80.6%,94.1%,0.893,respectively.The RD values of CPN in GBS group and healthy volunteers were 0.0008±0.0002,0.0011±0.0003,respectively.The difference was statistically significant(p<0.05).Using 0.0008 as the threshold,which was obtained by ROC analysis,the diagnostic sensitivity,specificity and AUC were 57.7%,100%,0.833,respectively.6.The correlation between DTI parameters and electrophysiology parametersThere was a decrease of MCV in all GBS group,31 TN and 29 CPV had decreased amplitude on electrophysiology.The FA,AD and RD values positively correlated with MCV and amplitude,and ADC values negatively correlated with MCV and amplitude(p<0.05).The FA values had the best correlation.7.The DTT manifestations of TN and CPN in GBS groupDTT showed that the TN and CPN in GBS patients were slender,the edges were unclear,and the number of fibers was decreased compared with the normal volunteers Part II:Application of DWIBS in the evaluation of TN and CPN injury with Gullain-Barre Syndrome and MR manifestations of denervated skeletal musclesObjectiveTo investigate the feasibility of diffusion weighted imaging with background suppression(DWIBS)sequence in displaying TN and CPN;to explore the value of DWIBS sequence in evaluating the injury of TN and CPN with GBS.To explore the MR manifestations of denervated skeletal muscles in T2WI SPAIR sequence.Materials and Methods1.Study populationTwenty GBS patients(13 men,7 women,age range 14 to 71 years,mean age:40 years)and sixeen healthy volunteers were enrolled in this study.Thirty six lower extremities of 20 GBS patients and thirty two lower extremities of 16 healthy volunteers were examined.2.MR examinationAll subjects were examined using a commercially available clinical 3T MRI system(Achieva,Philips Healthcare,Best,the Netherlands).An eight-channel knee coil was employed at the knee level.The subjects were placed supine with feet-first position and unilateral imaging.Axial and coronal SPAIR:TR/TE 3000/55ms,FOV 120 mmx 120mm,slice thickness 3.0mm,overlap 0.1mm,NEX 2.DWIBS:TR/TE 9000/86ms,FOV 10O×100mm2,acquisition matrix 75x72,overlap 0,slice thickness 3.0mm,slices 30,EPI factor 41,half-fourier factor 0.795,NEX 2,sensitivity-encoding factor 2.3,b value 800s/mm2.3.Post-processing and analysis of DWIBS imagesFull-volume maximum intensity projection(MIP)reconstructions were generated to produce three-dimensional(3D)images and Volume editing tool was manually used to remove superimposed structures that showed similar hyperintensity.The certainty of identifying the nerves of GBS group and healthy volunteers were analyzed by 2 experienced neuroradiologists.The imaging quality was evaluated using a 4-point grading scale:4,excellent(the entire nerve is clear and of good signal intensity);3,good(the entire nerve is relatively clear and of moderate signal intensity);2,moderate(the entire nerve is relatively clear and of week signal intensity);and 1,poor(the nerve is partially visible,difficult to identify or with obvious deformation).The MR manifestations of denervated muscles controlled by TN and CPN were analyzed by 2 experienced neuroradiologists blinded to each other.Diagnostic criteria included that the boundary of denervated muscles was not clear,T2 WI SPAIR high signal was seen in muscle tissue,muscle space or surrounding adipose tissue.If the diagnosis opinions were not uniform,the final result was agreed to decide by two observers.4.Statistical analysisInterobserver agreement for visualization scores of the TN and CPN in DWIBS was calculated using the kappa statistic.The level of agreement was defined as follows:values of 0-0.40,poor agreement;0.41-0.60,moderate agreement;0.61-0.80,good agreement;and 0.81-1.00,excellent agreement.Statistical analysis was performed using SPSS software(version 19).Results1.DWIBS manifestations of the TN and CPN in GBS group and healthy volunteersDWIBS showed that all the 32 TN and CPN of 16 healthy volunteers were clear,sharp edges and of good signal intensity.The scores of the two observers were 4.0±0.0,4.0±0.0,respectively.Interobserver agreement was excellent(K=1).DWIBS manifestations of the 36 TN in GBS group were as follows:5 were of scale 1,10 were of scale 2,15 were of scale 3,and 6 were of scale 4;The scores of the two observers were 3.65±0.54,3.63±0.47,respectively.Interobserver agreement was good(K=0.879).DWIBS manifestations of the 36 CPN in GBS group were as follows:6 were of scale 1,11 were of scale 2,14 were of scale 3,and 5 were of scale 4;The scores of the two observers were 3.58±0.61,3.56±0.54,respectively.Interobserver agreement was good(K=0.863).2.The T2WI SPAIR manifestations of denervated skeletal muscles in GBS group In the denervated skeletal muscles of 20 GBS patients(36 knees),11 cases of anterior,lateral and posterior group muscles were involved,5 cases of anterior group muscles were involved,6 cases of posterior group muscles were involved,6 cases of lateral group muscles were involved,8 cases of anterior,lateral and posterior group muscles were not involved,respectively.The MR features of denervated skeletal muscles comprehend hyperintensity on T2WI SPAIR sequence and the instinct boundary,muscle atrophy wse seen in 1case only.Conclusions(1)The DTI quantitative parameters(ADC,FA,AD and RD values)of TN and CPN in healthy volunteers and GBS group were significantly different.DTI could effectively evaluate the peripheral nerve injuries in GBS patients.(2)There was a good correlation between the DTI parameters of TN and CPN with electrophysiological parameters in GBS patients.ADC values negatively correlated with MCV and motor nerve conduction amplitude,and the FA,AD and RD values positively correlated with MCV and motor nerve conduction amplitude.The FA values had the best correlation.(3)DTT can clearly show the shape and course of the TN and CPN in healthy volunteers and GBS patients,which can also reflect the damage of peripheral nerve in GBS.(4)DWIBS can clearly show the TN and CPN in three dimensions,and has a relatively high contrast,and reflect the injury of peripheral nerve in GBS patients to some extent.(5)T2WI SPAIR sequence can clearly show the morphology and signal changes of skeletal muscle in the dominant region.(6)MRN can reflect the demyelination,axonal injury changes of peripheral nerve in GBS from cellular and molecular level,and can evaluate objectively,non-invasively,quantitatively.The combination of multiple MRN techniques allows us to understand the biological characteristics of GBS peripheral nerve injuries from different perspectives,which can provide more adequate evidence of TN and CPN injury in GBS patients and has important clinical significance. |
PDF Full Text Request